Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

"We don’t count, we’re just not there" : using feminist action research to explore the relationship between… Reid, Colleen 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2002-750728.pdf [ 15.88MB ]
JSON: 831-1.0076911.json
JSON-LD: 831-1.0076911-ld.json
RDF/XML (Pretty): 831-1.0076911-rdf.xml
RDF/JSON: 831-1.0076911-rdf.json
Turtle: 831-1.0076911-turtle.txt
N-Triples: 831-1.0076911-rdf-ntriples.txt
Original Record: 831-1.0076911-source.json
Full Text

Full Text

"We don't count, we're just not there": Using feminist action research to explore the relationship between exclusion, poverty and women's health by Colleen J. Reid B.PHE Queen ' s University at Kingston (1993) B.A. Queen ' s University at Kingston (1993) M.A. The University of British Co lumb ia (1997) A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF D O C T O R OF PHILOSOPHY in INTERDISCIPLINARY STUDIES (Health Promotion, Women's Studies, Educational Studies) We a c c e p t this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 2002 © Co l leen J. Reid, 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract One of the greatest social injustices is that people who are marginalized experience more illnesses, disability, and shorter lives than those who are more affluent (Benzeval, Judge, and Whitehead, 1995). In this dissertation I critique the notion that health is affected by poverty through primarily material factors. In fact, poor women are systematically excluded from resources and opportunities to pursue their health. This feminist action research project addressed how poverty and exclusion influenced poor women's health, examined how a group of women negotiated their experiences of poverty and health, and developed action strategies to address their shared concerns. For 1 V2 years I worked with a group of 30 poor women and gathered qualitative data from 15 meetings, 32 interviews, and 30 sets of fieldnotes. The women lived in material deprivation and could not afford the most basic living necessities. They felt stereotyped, excluded, and invisible in their every day lives. The stereotype of the "welfare recipient" fueled institutional stigmatization and surveillance. Welfare, health care, and community recreation workers were threatening, withheld important information, and limited the women's access to services through chscriminatory practices and policies. The women had limited access to health-promoting resources, and their interactions with authorities were shaming which negatively influenced their psychosocial health through stress, depression, low self-esteem, and anger. Services that were meant to help them labelled them as poor and hurniliated them. The women's shame, material scarcity, and limited access to resources engendered feelings of lack of control and hopelessness and influenced their health. The women's varied discourses of poverty and health reflected attempts at finding legitimacy in a society that systematically excluded and de-legitimized them. Through their conversations and our feminist action research work together, they uncovered legitimate identities within experiences of poverty and ill-health and advocated action and social change. They cited a "livable" income, accessible health-promoting resources, and redressing stigmatizing practices and policies as changes required to improve their health. These findings confirmed that the social determinants of health must be reframed to better understand the effects of exclusion on poor women's health and that inclusion, respect, and dignity are fundamental conditions for promoting health. Table of Contents Abstract ii Table of Contents iii List of Figures and Tables vii Acknowledgements viii Prologue ix 1 Introduction 1 Global Trends: Poverty and Exclusion 1 Women's Health and Social Justice 3 The Role of Feminist Action Research 5 Representational Challenges: My Voice as the Researcher 7 Overview 9 2 Poverty, Exclusion, and Women's Health: Using Feminist Action Research to Problematize and Address Poor Women's Health 11 The Dominant Discourses of Poverty and Health 12 The Dominant Discourse of Poverty ; 12 The Dominant Discourses of Health 14 Understanding Poverty, Inequity, and Women's Health 18 The Gradient of Health 19 Psychosocial Health and Biological Expressions of Social Inequality 21 The Health Equity Field: Conceptual Limitations and Oversights 23 Gender-Blind Conceptualizations and Measurements of Socioeconomic Status 24 Narrow Explanatory Frameworks 26 Dearth of Qualitative Studies 28 Limited Conceptuakzations and Measurements of Women's Health 28 What is Women's Health? 29 Poor and Unhealthy Women as "Other": A Theory of Exclusion 31 Unpacking the "Other": Provisional Identities and Women's Diversity 35 Feminist Action Research: A Response to Poor Women's Exclusion 38 Participatory Research 39 Feminist Action Research 40 Praxis: Developing A Feminist Action Research Framework 43 A Feminist Action Research Framework 44 The Marginalization and Co-optation of Feminist Action Research Ideals 48 3 Research Context and Methods 57 The Research Context: In Collaboration with WO AW 58 A Brief History of "Women Organizing Activities for Women" (WOAW) 58 Important WO AW Phraseology 62 WO AW Members: The People Involved in WO AW 62 Layers of the WO AW Organization 64 Conducting FAR Collaboratively 68 Data Gathering 70 Research Questions 70 The Women's Involvement as Research Participants 72 Chronology and Overview of Data Collection 73 Gaining Access and Building Trust: September 1999 - February 2000 73 Intensive Data Collection: March 2000 - June 2001 74 Continued Researcher Involvement and Commitment: July 2001 to the Present 76 Data Sources 77 The Research Team 77 One-On-One Interviews 80 Participant Observation and Fieldnotes 84 Data Analysis: "Sense Making" 85 The Analysis Process 85 The Realities of Collaborative Data Collection and Analysis 86 Power and Representation: Attempts at Deliberate Reflexivity 89 Withdrawal From the Field 92 Honoraria Payments 93 Anonymity and Confidentiality 94 Legitimation and Trustworthiness 94 Crystallization 95 Credibility 96 Transferability 97 Adas.ti 4.1: A Tmstworthiness Tool 98 Writing '. 99 4 The Women's Stories 103 The Women's Lives and Experiences 105 Aged to Perfection Subgroup 105 Rene 105 Trina 107 Wanda 109 CoPoMo Subgroup 112 Alexa 112 Katharine 113 PoCo WO AW Subgroup 116 Joanne 116 Kelly 117 Virginia Dawn 121 SWCo Subgroup 123 Elizabeth 123 Maey 125 Susan 126 Willow 130 Common and Divergent Themes 133 Age 133 Living in Poverty 134 Past and Future Employment 135 Education 136 Domestic and Parenting Status 136 Race, Ethnicity, and Sexual Orientation 137 Health .' 139 5 Poor Women's Exclusion and Health 142 Experiences Living In Poverty: Cultural, Institutional, and Material Exclusion 142 Cultural Exclusion: "They label you, they typecast you" 143 iv Stereotyping: Welfare Single Mothers 144 Invisibility: Women as the "Unseen Poor" 145 Simultaneous Stereotyping and Invisibility 147 Experiences of Institutional Exclusion: Victimizing Policies and Practices 149 The Welfare System 149 The Health Care System 153 Community Recreation Departments 156 Material Exclusion 160 Housing 160 Parenting Costs 161 Transportation 161 Exclusion and Poor Women's Health 163 Psychosocial Health Problems as "Shame Markers" 164 The Cost of Good Health 171 The Stress and Depression of Material Scarcity 173 Unhealthy Behaviours 175 Smoking and Other Addictions 175 Unhealthy and Disordered Eating 176 Physical Inactivity 178 Exclusion, Inequality, and Women's Health 180 6 Negotiating The "Other" Identity: Legitimacy, Power, and Discourses of Poverty and Health 184 The Powerless Victim Discourse 185 "Bad Apples" Are Disempowering 186 Welfare Workers and the Industry of Poverty Are Disempowering 187 Stress and Ill-Health Happen 190 The Legitimacy and Enablement Discourse 192 Biomedical Health Problems Are Legitimate 192 Disability Is Legitimate Dependency 195 The Individual Work Ethic Discourse 198 Good Mothers Choose Dependency 199 Poverty and Health Are Individual Responsibilities 200 Strong Individuals Can Survive Poverty 204 The Critical and Collectivist Discourse 206 Negotiating the "Other" Identity: From Individuality to Empowerment? 208 7 Lofty Ideals and Lived Experiences: The Research Team as Feminist Action Research 213 The Challenges of Inclusion: Voice, Learning, and Emotions 214 "Voice" and Sharing the Same Problems 214 Learning, Problem-Solving, and Networking 219 The Emotionality of The Research Team 221 The Research Team: Fostering Inclusion and Exclusion 222 Participation: Work and Conflict 224 Workers and Non-Workers 224 Structuring Work 227 Symbols of Work 228 Eroding Group Process and Conflict 229 Participation, Work, and Feminist Action Research 230 Individual and Collective Actions Towards Social Change 232 v The Women: Individual Actions 233 The Research Team: Collective Actions 235 Powerfulness and Powerlessness in the Work of Collective Action 239 The Onus for Action 243 "The value is being together but the goal is to change the world" : Did Social Change Happen? 245 Reflections From the Field 248 My Role as the Feminist Action Researcher 248 The Researcher as a "Researcher" 249 The Researcher as a "Facilitator" 250 The Researcher as a "Research Team Member" 252 Confronting Reflexivity, Power, and Emotions 254 The Research Team as Feminist Action Research 259 8 Women's Health and Social Justice: Implications, Recommendations, and Conclusions 262 Implications: Policy and Practice 264 Support community development projects like WO A W 265 Advocate participatory policy development 266 Uncover alternative sites for social justice 266 Redress material deprivation 267 Promote redistribution and equity 268 Recognize gender and women's diversity 269 Politicize the social determinants of health 270 Ideas for Future Research Projects 271 A Vision of Health and Social Justice 273 Epilogue 2 7 4 References 279 Appendices Appendix A - Overview of SSHRC Project 295 Appendix B - Detailed Overview of Research Team Meetings 297 Appendix C - Letter of Initial Contact and Informed Consent Form 299 Appendix D - Schedule #1 Interview Questions March 2000 301 Appendix E - Schedule #2 Interview Questions June 2000 302 Appendix F - Interview Face Sheet 303 Appendix G - Fieldnote Guidelines for Observations of Research Team Meetings 305 Appendix H - Code Book 306 v i List of Figures and Tables Figure 1 - The Social Determinants of Health 22 Figure 2 - The WOAW Organization 67 Table 1 - The Women's Subgroup Affiliations 66 Table 2 - Overview of SSHRC and Dissertation Data 71 Table 3 — Key for Reading Quotations 72 Table 4 - Overview: WOAW Women's Involvement in Dissertation 73 Table 5 - Research Team Meetings (March 2000 - June 2001) 79 Table 6 - Data from the Research Team Meetings 80 Table 7 - One-on-one Interviews (March - September 2000) 83 Table 8 - Data from the Interviews 84 Table 9 — Data from the Fieldnotes 85 Table 10 - Overview: Dissertation Data 85 Table 11 - Profile of Individual Research Participants (n=20) 104 Table 12- The Women's Ages 133 Table 13 - The Women's Poverty Profile 135 Table 14 - The Women's Education 136 Table 15 - The Women's Domestic and Parenting Statuses 137 Table 16 - The Women's Race and Ethnicity 139 Table 17 - The Women's Health: "Do you consider yourself to be healthy?" 139 Table 18 - Self-Reported Health Problems 140 Table 19 - Examples of Individual Actions and Problem-Solving 235 Table 20 - Examples of WOAW and Research Team Collective Actions 237 vii Acknowledgements This dissertation would not have been possible without the participation of the following "WOAW women" - Alexa, Arlene, Caroline, Cassie, Christine, Cynthia, Elizabeth, Gloria, Helen, Joanne, Julie, Katharine, Kelly, Maey, Martha, Rene, Susan, Theresa, Trina, Virginia Dawn, Wanda, and Willow. I learned and grew from our work together; they taught me what it means to be truly courageous, strong, and committed. Working with them leaves me hopeful that social change is possible. I would also like to thank the WOAW organization and the community service providers who warmly welcomed me — Louise Hara, Carla Alexander, Brenda Devine, Jim LaCroix, Gwen Wong, Andrea MacDonald, Diane Lego, Pam Bailey, and Bonnie Ste. Croix. I am inspired by their tenacity and dedication to their work. I would especially like to thank Louise Hara for helping me to better articulate my research interests and determine the need for the Research Team. Dr. Allison Tom, my research supervisor, is a remarkable mentor and friend. With her uncanny ability to get to the core of difficult research issues and think deeply and analytically she enabled me to find the clarity and direction I needed. Allison shared my enthusiasm for my research while challenging me with humility and compassion. I thank Allison for enabling me to find my voice as a feminist researcher. Dr. Wendy Frisby, my Master's advisor and Ph.D. committee member, provided me with the opportunity to work with WOAW. We have shared the ups and downs of community-based research and I have learned from her insight and commitment to the ideals of feminist action research. I would also like to thank Dr. Carol Herbert, who despite moving to London Ontario remained an invaluable and active committee member, and to Dr. Penny Gurstein, who supported and encouraged me all along the way. There were two groups of students who greatiy enhanced my doctoral work at UBC — the W O A W Working Group and Ournet. The WOAW Working Group, including Pamela Ponic, Sydney Millar, and Beth Partington, has been my sustenance for the last three years. Our discussions and support for one another transformed my Ph.D. from an often lonely and isolating experience to one of connectedness and sharing. I would particularly like to thank Pam for her insight and intellect. Together we have dissected challenging research issues, written papers and presentations, and openly shared our own biases and assumptions. Ournet provided a rich learning environment that stimulated me both intelllectually and emotionally. In particular I'd like to thank Marina Niks, Lynette Harper, and Johanna Kuyvenhoven for the conversations, passion, and friendships we have shared. Finally, I thank my family. My parents, Heidi Reid and Gordon Reid, have always mstilled in me a sense of confidence and possibility. They encouraged me to pursue my passions and to never compromise my interests for the sake of conventions. My sister, Heather Reid, has expanded my world vision through helping me connect authentically with my work. My life partner, Rod Tukker, has been a pillar of support and encouragement every step of the way. He has facilitated this journey in innumerous ways. I feel such gratitude having you in my life — thank you. vm Prologue When I was eighteen, a friend of mine worked for Ottawa Parks and Recreation as a wading pool lifeguard. She told me that the job was fun and social and that as a lifeguard you worked outside all day and programmed activities for children in the local area. I was hired and stationed at a pool in one of the lowest income areas in Ottawa. Over that summer I witnessed unbelievable stories of deprivation and hopelessness — a four year old child at the pool for 10 hours a day with no food, teenagers selling merchandise they had stolen the day before, a six year old girl babysitting her two year old brother, a twelve year old girl telling me that she wanted to be a "hooker" when she grew up. Almost every day someone asked me for money and my lunch. I saw parents who were so impoverished that they could not provide a secure home for their children; most of them lived in violence and scarcity in what the local residents called "the foster farm." What truly amazed me, however, was not the extent of the deprivation, but how, despite living only 10 minutes apart, we had drastically different trajectories of every day life. My security and sense of reality, that were nesded in my suburban middle-class upbringing, became unhinged; poverty and my own privilege were now visible to me. Watching the news coverage of the Montreal Massacre in 1989 was another experience that profoundly shifted my worldview. As I watched the paramedics carry the women's bodies out of the engineering building, I saw parents and friends approach the scene, confused and panicked. I remember seeing a bewildered boyfriend learn that his girlfriend had been murdered. I was stunned that such an act of violence, injustice, and hate could occur, and realized that this event could just as easily have happened in one of my classes at Queen's University. I could have been one of the women being carried out on the stretcher. I now recognize how these experiences — my work at the wading pool and watching the ix Montreal Massacre — had a formative influence on what I am pursuing today. They led me to identify with the feminist movement and to uncover my passion for examiriing social injustices, reflecting on my own assumptions and privilege, and understanding women's roles and experiences. I now believe that through remaining compassionate and humble every person has a role to play in the pursuit of social justice. Fundamentally, this dissertation is my attempt at taking part in the work of social justice. Justice to me is a warm spirit, born of tolerance and wisdom, present everywhere, ready to serve the highest purposes of rational wo/man. To seek to create the just society must be amongst the highest of those human purposes. Because we are mortal and imperfect, it is a task we will never finish; no government or society ever will. But from our honest and ceaseless effort, we will draw strength and inspiration, we will discover new and better values, we will achieve an unprecedented level of human consciousness. On the never-ending road to perfect justice we will, in other words, succeed in creating the most humane and compassionate society possible. Pierre Elliott Trudeau x Introduction The economy is stacked this way and the social situation is stacked this way and the problems you're having are certainly not of your own making. And when you understand what's working against you out there, you feel much stronger to face up to it You're a human being and you deserve dignity and you should have some rights, when you're not fully aware of what the society... it's our food, it's our politics, it's our institutions (Helen, R T M June 7, 2000). Global Trends: Poverty and Exclusion In the last sixty years major transformations in economic, social, and political systems have occurred. Arguments for social progress and increased economic efficiency have rationalized the deregulation of global markets, the development of supra-national institutions, and the decreasing role of government. These globalization1 trends support a reduced role of government because of the purportedly "self-regulating" nature of global economies and the threat of government intervention to economic processes. Governments increasingly support corporate interests and sanction industry relocations to places where labour costs are lower, unions are non-existent, and state regulations are less expensive (Fine and Weis, 1998). Although corporate-centered trade has displaced jobs, capital-intensive investment remains in North America. With the reduced role of government, job displacement, and accumulation of capital, the wealthy become wealthier and income distribution widens. Not only have globalization trends meant that fewer jobs are available to poor and working-class people, but Canada's welfare state has also been affected. "Amid globalization, the increasingly transnational activities of the private sector circumvent and even undermine the 1 capacity of individual countries to ensure social protection and welfare. Countries competing for foreign investment may reduce their social policy standards to be more economically attractive" (Bettcher and Lee, 2002, p.10). As a consequence of political and economic trends that favour corporate interests, in the early 1980s the welfare state entered a fiscal crisis and lost legitimacy. The principle of universality in government services was blamed for excessive spending, inordinate bureaucratization, and halted economic progress (Harvey, 2001). The diminution of manufactxiring-based employment, twinned with the shredciing of the public safety net, produced conditions that exacerbated the growing inequality between the rich and the poor (Fine and Weis, 1998). Indeed, "the legacy of the 20 th century is the cynical defeat by capital and the state of social equality" (Fine and Weis, 1998, p.258). An important aspect of the legacy of the 20* century is the spilling over of the self-interested individuaHsm of the marketplace into other areas of social life (Wilkinson, 1996). At the centre of the concept of individuaUsm is the practical separation of each person's interests and identity from those of others. According to Wilkinson (1996), individuaHsm is most fundamentally expressed by the role of cash in a market economy as we earn and spend our "living," opposed as buyers and sellers in the marketplace. Everyone is dependent on their individual incomes, and any acknowledgement of others' needs is perceived to endanger the sufficiency of one's income and security to meet one's needs fWilkinson, 1996). With the loss of community and rising individualism, people tend to assume that their position in society is a reflection of their innate worth and that the poor are the principal authors of their own fate (Harvey, 2001). The logic of possessive individualism that is engendered with globalization limits some people's rights and their ability to fully function in society. The notion of participation affirms the human right of persons to contribute to decisions that affect them, and is also fundamental to human flourishing. Those who conform to the market-driven ideals of individualism are included in society and can participate, while those who fall outside of the criteria for inclusion and participation, because they are single mothers, sick, disabled, or poor — because they are dependent — are excluded (Reason, 1998). Exclusion involves disintegration from common cultural processes, lack of participation in societal activities, alienation from decision-making and civic participation, and barriers to employment and material resources (Raphael, 2001). Individualism has changed the nature of public life such that human interaction has become dominated by the asocial values of the market (Wilkinson, 1996). These asocial and individuaHstic values that drive the economy and human interaction legitimize exclusion and the withholding or denial of an individual's right and need to participate equitably in society. Women's Health and Social Justice In almost any developed society people lower down the social scale may have death rates two to four times higher than those nearer the top; more egalitarian societies, that is societies with smaller differences in income between rich and poor, tend to have better health (Wilkinson, 1996). As health and society are closely related, learning about society can teach us about health, and learning about health can teach us about society. Increasing poverty and income inequality are mirrored by increasing health inequaUties (Wilkinson, 1996; Shaw, Dorling, and Smith, 2000). The social and economic structure of society, especially low income, income inequaHty, mscrimination, and social exclusion, can be seen as the ultimate determinants, the "causes of the causes," of disease and death (Deaton, 2002). In Canada, there is rising poverty, inequality, and unemployment alongside reductions in social, health care, and community services. Socio-demographic changes, aging populations, and rising numbers of lone parents also contribute to social inequities and a greater proportion of people living in poverty. A purpose of this dissertation was to better understand how a group of 3 women on low income2 understood and experienced poverty and health. The societal trends of increasing individuation, reduced social cohesion and community, and the exclusion of the poor suggest that better understanding the relationship between exclusion and women's health is relevant and timely. Social inequities' assault on health is indeed too significant to be ignored. Health inequities are also significant because health is widely recognized as a fundamental right of citizenship "the enjoyment of the highest attainable standard of health is one of the fundamental rights of every human being without distinction of race, religion, political belief, economics, or social condition" (WHO, 1948; cited in Hankivsky, 1999). In that vein, health is a social justice issue — "it is one of the greatest of contemporary social injustices that people who live in the most disadvantaged circumstances have more illnesses, more disability, and shorter lives than those who are more affluent" (Benzeval et al., 1995, p.l). According to Young (1990), social justice concerns the degree to which a society contains and supports the conditions necessary for all individuals to exercise capacities, express experiences, and participate in determining actions. Social justice requires not the melting away of difference, but the promotion and respect for group differences without oppression (Young, 1990). If there is not a sense of social justice in society, then the legitimacy of social institutions is fundamentally weakened and the moral community which makes social life coherent is lacking (Wilkinson, 1996). Numerous international platforms explicidy focus on women's right to health as an integral component of human rights protection and promotion.3 International conventions, documents, and platforms obligate the global community, mcluding Canada, to take concrete action to eliminate all forms of discrimination against women. Importantly, it is not enough to state or recognize human rights, rather, conditions must exist in which they can be exercised and realized (Hankivsky, 1999). Health can be seen as a social justice issue since social justice is concerned with creating the 4 opportunities for attaining full health potential and reducing health inequities (Hankivsky, 1999). Equity refers to conditions largely out of individuals' control that create unjust differentials in health.4 Lack of power at the individual, community, and societal levels is a major risk factor for poor health. Empowering the disadvantaged — or disempowering those who use their privilege to benefit themselves at the expense of the wellbeing of the community — is an important tool for health promotion. Protecting and restoring health involves a social justice ethic based on collective action and fair play that respects individual rights and experiences (Wallerstein and Freudenberg, 1998). The Role of Feminist Action Research The social transformations arising from globalization, mounting rates of and concerns about exclusion, and the rise of poverty and income polarization, are reasons for retbinking how we conceptualize and study poor women's health (Harvey, 2001). Social justice is not only a way of seeing the world, it can also inform how research is conceptualized and conducted. It is possible to locate issues of social justice front and centre on research agendas through carefully appreciating the pohtical nature of research (Poland, 1998). To develop theories and methods to examine social injustice profoundly influences how we engage with marginalized groups and conceive our role as researchers (Wilkinson, 1996). Feminist action research (FAR) is a conceptual and methodological research framework that is fundamentally about exploring and pursuing opportunities for social justice. Feminist action researchers expHcidy center women's experience and diversity in practical and explanatory frameworks (Maguire, 1987) and apply the principles of inclusion, participation, action, social change, and reflexivity throughout the research process. Feminists envision justice as according respect and participation in decision-making to those who are dependent (Young, 1990). FAR's methods aim to open horizons of discussion and to create spaces for collective reflection in 5 which new descriptions and analyses of important situations may be developed as the basis for new actions (Greenwood and Levin, 1998). The second purpose of this dissertation was, through applying FAR principles, to examine the process of working collaboratively towards social change. As a project manager and doctoral researcher I worked with a collective organization named "Women Organizing Activities for Women" (WOAW). WOAW was a partoership between university-based researchers, community service providers, and women on low income that explored and addressed a group of marginalized women's barriers to meaningful participation in their communities. The complex structure of W O A W enabled different kinds of participation for all of the members. Most of the women on low income participated in WOAW's activities, including community recreation, workshops, and social activities. Many WOAW members were involved in WOAW's organizing, such as attending meetings, planning workshops, and coordinating activities. Research was the third component of WOAW. The Social Sciences and Humanities Research Council of Canada (SSHRC) provided three years of funding to examine women on low income's health problems and different forms of community organizing; over half of WOAW members were involved in the SSHRC research. As researchers we, a group of professors and graduate students at the University of British Columbia (UBC), conducted interviews and focus groups and recorded fieldnotes for most WOAW meetings. For a year and a half a group of the women on low income and I worked together as the Research Team. It was with the members of the Research Team that I conducted the research for this dissertation — the one-on-one interviews, group meetings, participant observations, and fieldnotes. I posed three research questions: (1) How did a group of women on low income understand and negotiate their experiences of poverty, exclusion, and health, both individually and collectively? (2) From the perspectives of a group of women on low income, what was the 6 connection between poverty and health? In what ways did they believe that the experiences of exclusion, stereotyping, and invisibility affected their health? and (3) How did a group of women on low income collaboratively identify, organize, and enact action towards social change to address their shared concerns? In framing these research questions I attempted to avoid dichotomizing individual volition and societal determinism. In other words, to entirely focus on the individual risks privatizing by rendering personal the social and economic underpinnings of poverty and powerlessness, yet focusing only on structural issues may ignore "the immediate pains and personal woundings of the powerless and people in crisis" (Labonte, 1994, p. 259). I hoped to portray the complexity of the women's personal experiences, strengths, and abilities, and the structural constraints they faced. The process of exploring the women's shared constraints and experiences of poverty and health was guided by feminist action research. An examination of ideals and realities of inclusion, participation, action, and social change underpins the third research question. Representational Challenges: My Voice as the Researcher For a novice feminist action researcher, working with W O A W as a project manager and doctoral researcher was a unique and formative experience for which I am deeply grateful. Importandy, the proceeding analysis reflects how I experienced W O A W and the Research Team and is by no means a complete picture of all that occurred with this complex organization and its diverse members. As Wendell (1996) wrote, "I do not imagine that my own worldview is complete or even very accurate" (p.104). My analysis is indeed only a partial picture that is conveyed through the lenses of my personal experiences, assumptions, and worldview. In this dissertation I position myself as a feminist conducting feminist action research. In the "rendered accounts" that follow I have attempted to hold myself accountable both to the research participants and to myself for my critical analysis and responsible use of power (Ristock 7 and Pennell, 1996). This involved significant representational challenges. I am a white, middle-class, heterosexual, educated woman speaking about the "Other" 5 — in this case a group of marginalized, stereotyped, and poor women. Under assumptions of finding a less exploitative approach to research, feminists previously sought counter-practices of knowing such as articulating personal voices and practices of co-writing. However, no longer can feminists assume "innocence" in a desire to give the voiceless a voice. Judith Stacey (1988) interrupted this drive to innocent knowing with focusing on the inescapable power imbalances of inquiry situations, feminist or not. She suggested that by setting itself up as better intentioned, feminist research risked even greater violation of the researched than the more distanced objectivity of conventional research methods (Lather, 2001). My representational efforts did not assume a cure or a solution, nor did I wish to join "the university rescue mission in search of the voiceless" (Visweswaran, 1994, p.69; cited in Lather, 2001). It is necessary to focus on the limits of our knowing and to also acknowledge that texts are inevitably and always a site of failures of representation (Lather, 2001). What remains from these feminist dilemmas is a necessary tension between the desire to know and the limits of representation that enabled me to question my authority without paralysis. My research was expliciuy mterdisciplinary and involved a blending of Uteratures from different voices, languages, research methodologies, and conceptual frameworks. In this text there are multiple approaches that endorse complexity, partial truths, and multiple subjectivities (Lather, 2001). Metaphorically, this dissertation can be envisioned as a crystal, with each chapter a side of the crystal that iUuminates an aspect of the women's lives. In approaching each chapter as a partial story or piece of the whole crystal, I attempted to layer complexity, foreground problems, and displace the privileged fixed position from which I interrogated and wrote about the researched (Lather, 2001). Each chapter has a different perspective, voice, and body of 8 literature that informed it, and I positioned the women's experiences alongside my theoretical analyses. In my representational efforts I "both got in and out of the way " (Lather, 2001, p.215). Overview This dissertation has eight chapters. In chapter 2, I review Mteratures that range from social epidemiology, women's health, discourses of poverty and health, to feminist action research. I argue that a theory of cultural, institutional, and material exclusion advances understandings of poor women's health and that a feminist action research framework can facilitate such understandings while moving towards collective action. The research context, "Women Organizing Activities for Women," and the research methods are described in chapter 3. In this chapter I provide detailed descriptions of the research methods and analysis, my influence as the researcher, and the challenges of reflexivity, representation, and trustworthiness. Chapters 4, 5, 6, and 7 comprise the analytic sides of the crystal. Chapter 4 is an overview of the women's lives and experiences. This side of the crystal is data-rich and descriptive. On chapter 5's side of the crystal, I explore how the theory of exclusion arose from the data and argue that exclusion has a serious influence on the women's health in terms of shame, material deprivation, psychosocial health conditions, and health behaviours. In chapter 6, I interpret the data on a discursive level by making the women's discourses themselves the object of study. I present the ways the women negotiated their lives and experiences and sought legitimacy and power in a society that systematically excluded and "othered" them. Chapter 7's side of the crystal provides an analysis of the Research Team through drawing on the principles of feminist action research that were described in chapter 2 — inclusion, participation, action, social change, and researcher reflexivity. I explore my role as the feminist action researcher and my emotional reactions to the research process. The exploration of my role, experiences, and emotions is at the end of this chapter because I did not want to foreground the women's experiences with my own 9 reflections. Finally, in chapter 8 I argue that health is a social justice issue and suggest pohcy and research recommendations. In this dissertation I attempt to provide as complete a picture as possible of a group of poor women's exclusion and health and their experiences in a feminist action research project. Yet, "fieldwork is a morally ambiguous enterprise. The question for me is how to show respect for the people I write about, given the impossibility of complete sincerity at every moment" (Duneier, 1999, p.336). I wrote to inform rather than to impress and to openly portray the women's experiences and myself within and throughout their experiences (Duneier, 1999). In striving to write with compassion, humility, aclmiration, and outrage, I "practiced presence." Being present invites us to set aside judgement yet still retain discernment, to set aside sympathy yet hold to empathy, to set aside power yet be responsible for influence. It invites us to bring head, heart, and body into this moment (Woodman, 2001). Notes 1 Globalization can be defined as a set of processes leading to the creation of a world as a single entity, relatively undivided by national borders or other types of boundaries such as cultural or economic. Globalization contributes to intensified human interaction in a wide range of spheres (economic, political, social, environmental) and across three types of boundaries — spatial, temporal, and cognitive — that have hitherto separated individuals and societies. It also influences the timeframe of human interaction and thought processes (Bettcher and Lee, 2002). 2 The research participants used the terms "poor women," "low-income women," "women on low income," "women with limited resources," "women facing barriers," and "isolated women" alternatively. There was tension and disagreement around the use of terms such as "poor" and "low income." Given that there was no consensus among the women, both "women on low income" and "poor women" are used throughout this dissertation. 3 According to the United Nations Social and Economic Council (1999), "the realization by women of their right to the enjoyment of the highest attainable standard of physical and mental health is an integral part of the full realization by them of all human rights, and that the human rights of women and the girl child are an inalienable, integral, and indivisible part of universal human rights" (cited in Hankivsky, 1999). 4 "If you live longer than I do or if you suffer from less sickness and disability, our health status is unequal. There is inequality between us, but not necessarily inequity. The difference may not result from our living conditions which may be essentially the same, but from accidents, genetics, and life choices. If however, the differences in our health status result from different living conditions, mine being less satisfactory than yours, a question of inequity arises. I may have less access to nutritious foods, difficulty in finding decent housing or high-quality health care sensitive to my particular needs. My income may be lower, and my work stressful and demoralizing, punctuated by frequent periods of prolonged unemployment. In this case, inequalities in health status are the result of inequities in life" (Draper, 1989; cited in Hankivsky, 1999). 5 The term "Other," as defined in chapter 2, is used throughout the dissertation. I distinguish it from common usages of the word other by capitalizing it - Other - and, when necessary, marking it with quotations — "Other." At times I have also used the terms "othering" and being "othered" to capture the process of being designated the Other. 10 Poverty, Exclusion, and Women's Health: Using Feminist Action Research to Problematize and Address Poor Women's Health Having no money is a prison in itself. You can go nowhere. The only tiling you can do is take a walk. You have no power, having no power you have no self-esteem I have to rebuild my life. How do you rebuild your life, how do you believe in yourself? Because we're brainwashed every day with these subliminal messages of people looking down on you 'cause you're on welfare. Al l this is very powerful. It's imprinted on the mind and you suffer it. Your mental plane, emotional plane, and everydiing suffers. So you're down (Katharine's interview, March 14, 2000). Although lack of access to effective medical care is likely to lead to unnecessary illness and suffering, there are inequaHties in the onset of new disease, which are not related to medical care, and inequaHties in mortality from diseases which are not always amenable to medical intervention (Marmot, 2000). In fact "to understand health or illness one must understand the social conditions in which health and illness are created, identified, defined, and continued" (Virchow, cited in Whiteford, 1996). The poor are often in bad health not from their own negligence, but from social conditions that systematically deprive them of resources and from historical conditions that isolate them (Whiteford, 1996). In this chapter I review literature from divergent fields of research mcluding discourses of poverty and health, social epidemiology, women's health, ferninist theory, and feminist action research. I review the dominant discourses of poverty and health in order to provide a broad landscape for locating poor women's experiences of health. I then argue that adopting a social determinants perspective is most suitable for studying poor women's health. Although the health 11 equity field's conceptualizations of the relationship between poverty and health have advanced understandings of the social determinants of health, they have not fully captured women on low income's lived experiences. As a consequence, I propose a theory of cultural, institutional, and material exclusion. Through proposing a theory of exclusion, I contribute to conceptualizations of poor women's health and the ways that women on low income negotiate their day-to-day lives. In the final section of this literature review I explain the genesis and principles of feminist action research (FAR), how the use of FAR can elucidate and address poor women's health, and the ways that it is a viable research framework for the work of collective action. The Dominant Discourses of Poverty and Health A discourse is a socially structured way of knowing that affects our subjectivity and social relations (Weedon, 1987). Discourses are sets of shared and often unconscious assumptions reflected in language that position people in relation to one another and frame knowledge (Ristock and Pennell, 1996). Discourses co-exist and compete with each other; some discourses are marginalized while others are authoritative and dominant (Fraser, 1992; Kelly, 1996; Batsleer and Humphries, 2000). Although societies contain a plurality of discourses and discursive sites and a multitude of positions and perspectives from which to speak (Fraser and Gordon, 1997), the dominant discourses of poverty and health reflect ideological agendas that are both historically specific and socially situated signifying practices. The Dominant Discourse of Poverty The dominant discourse of poverty is framed in terms of welfare dependency. As Fraser and Gordon (1997) argue, in the 16 th and 17 th centuries dependency was seen as a normal condition — a social relation associated with wage labour — and was not seen as an individual character trait. Independence was associated with privilege, superiority, increased political rights, and freedom 12 from labour. In the 18th and 19 th centuries dependency shifted and was framed as both a social relation and an individual character trait. Positive images of independence proliferated while dependence was seen as antithetical to citizenship. Any work associated with remuneration, including wage labour, became associated with independence, while people excluded from wage labour were personified as dependent. Increasingly, as can be seen in stereotypes of the "pauper" who lived on poor relief, character defects were believed to cause poverty (Fraser and Gordon, 1997). Currendy the dorninant discourse of poverty has two dimensions. The rhetoric of pauperism and the theory of the culture of poverty contend that poor and dependent people have something more than lack of money wrong with them. The flaws can be located in biology, psychology, upbringing, or neighborhood influence. The properties once ascribed to social relations become inherent character traits of individuals or groups, as if the social relations of dependency are absorbed into personality. The dominant discourse of poverty is also based on the economic premise of a "rational man" who faces choices in which welfare and work are both options. Whether poverty is a consequence of individual character flaws or notions of choosing or not choosing work, this discourse suggests that "dependents" have free will and control over their dependency (Fraser and Gordon, 1997). Today "dependency" is an ideological term that most often refers to the condition of poor women with children who maintain their families with neither a male breadwinner nor an adequate wage and who rely for economic support on the government. Social and economic problems are blamed on women rather than on the profit-oriented decisions of business and the state (Abramovitz, 1995). Naming the problems of poor families as dependency tends to make them appear to be individual problems (Fraser and Gordon, 1997). Since welfare dependency is an individual character flaw or a consequence of bad decision-making, women on welfare "are 13 depicted as ... embodying the necessity for welfare reform and of sitting at the heart of moral decay" (Fine and Weis, 1998, p.l). Thus poor women are thought to have no self-control, choose poverty and dependency, and legitimately cost hard-working taxpayers money. The dominant discourse of poverty moralizes welfare dependency as a consequence of character flaws or irresponsible decision-making. The notion of individual "free will" that underlies the dominant discourse of poverty bolsters the assumption that equahty of opportunity exists and individual merit determines outcomes (Fraser and Gordon, 1997). As individuation increases, interdependence is no longer understood as a necessary condition for autonomy and citizenship. "Dependence on welfare" is not something to expect at some point in our lives but a position to be avoided at all costs (Batsleer and Humphries, 2000). Consequentiy, dependency becomes increasingly megitimate and any self-evidendy "good" adult dependency in postindustrial society disappears. Al l dependency is suspect, and independence, which remains identified with wage labour, is enjoined upon everyone. Thus the worker becomes the universal social subject — everyone is expected to work and to be self-supporting. Any adult not perceived as a worker shoulders a heavier burden of self-justification (Fraser and Gordon, 1997). The disappearance of "good" or legitimate dependency in our increasingly individualistic society has heightened the stereotype of those dependent on B.C. Benefits1 and the welfare system itself. The welfare system is highly stigmatized, imbued with stereotypes, and portrayed as undermining motivation and accentuating an underclass mindset. With no legitimate dependency, welfare is questioned for promoting "out-of-wedlock" children and discouraging women from accepting jobs. Indeed, with the disappearance of "good" dependency, the stigma of poverty has reached unparalleled levels. The Dominant Discourses of Health "Health" provides the perfect metaphor for values that fundamentally structure our social 14 and cultural life (Crawford, 1984). We affirm each other and ourselves as well as allocate responsibility for failure or misfortune, through shared images of wellbeing. Just as the dominant discourse of poverty structures lived experiences for both the impoverished and the rich, the dependent and the independent, health has become a theme in day-to-day Hving. Thus the link between society and health not only serves to help us understand the cause of ill-health and disease, as health equity researchers2 argue, but it enables people to understand themselves in relation to others. The link between society and health has implications for how health is defined and understood. The ways that we conceptualize, speak and write about health are never just about health; they also function as repositories and mirrors of our ideas and beliefs about human nature and the nature of reality, as well as about the kind of society we imagine creating and how best to achieve it (Robertson, 1998). Unlike the dominant discourse of poverty, there are three dominant discourses of health — biomedical, individual responsibility, and social determinants discourses. Throughout the 19 th and 20 t h centuries the biomedical discourse prevailed. Health was seen as the absence of disease, and managing health and preventing disease gave rise to the power and predominance of medical institutions, personnel, and texts. Increased medicalized practices dominated expenditures on health and the treatment of disease, and the provision of universal health care, which was instituted in 1968, remains the cornerstone of the Canadian welfare state.3 As such the biomedical health discourse discredited the possibility of individual free will through locating the etiology and progression of ill-health, disease, and disability in biology and genetics and under the management of medical experts. Alongside the continued dominance of the biomedical discourse is the behavioural health discourse. Embedded within this discourse is the notion that poor health is an individual failing, inferring individual decision-making, a lack of education, and moral laxity on those who are unhealthy. Within this discourse health is an 15 important means for the middle class to structure its own class identity. Through mampulating this discourse the middle-class identifies itself as most capable of hard work, self-denial, asceticism, and clean living (Crawford, 1980). Everyone has free will, and the biological, structural, and systemic dimensions of health are disregarded. The biomedical discourse is exercised through "expert" knowledge and professional practices and the concomitant passivity of the receptacles of expert knowledge, the patient. The classist behavioural health discourse is sustained by the dominant middle-class which has the resources and time to adhere to particular health expectations. While the biomedical discourse of health places individual control and choice in the authority of medical experts, the behavioural discourse shames and views them as deviant and inferior non-conformists because they are unhealthy "on purpose" and have insufficient resolve to exercise more, to quit smoking, and so on (White, Young, and Gillett, 1995). Individuals, who for reasons beyond their control have neither the time nor the resources to apply to health management, are blamed for their failure to live up to their social responsibility for their own bodies (Labonte, 1982). Being ill means being guilty, the obese have let themselves go, smokers have no will-power, and non-aerobics are lazy (Becker, 1986). These dominant health discourses blame the victim or abstracted biological processes for ill-health, thus diverting attention from systemic, social, and structural factors that compromise the health of many. The social determinants of health discourse adopts a more ecological and sociological perspective and is tied to distal determinants of health including poverty, gender, race, ethnicity, and culture (O'Neill, Rootman, and Pederson, 1994). According to O'Neill, Rootman, and Pedersen (1994), the evolution towards a social or ecological view of health has enabled a broader conception of health to emerge. Krieger and Zierler (1995) contend that social, economic, and political conditions shape distributions and determinants of health, disease, and 16 wellbeing. Intervening in the public's health necessitates politics to reduce social and economic inequahties, curb environmental degradation, and increase options for social and individual action to improve health (Krieger and Zierler, 1995). Poverty's adverse effects of material deprivation and the home and workplace environment are seen to negatively affect health. As well, the social determinants of health discourse recognizes social class differences in health behaviour, and sets health behaviour alongside and in the context of the everyday circumstances in which people live and work. These factors are seen to shape and constrain what individuals can do to promote their health, setting limits on the choices they can make about their lifestyle (Graham, 1990). The dominant health discourses differentiy locate the power of the individual in exercising free will. The biomedical health discourse presents genetic and biological factors as the cause of ill-health and disability, while the social determinants of health discourse argues that social, pohtical, cultural, and economic factors cause the vast majority of disease and disability. In both discourses the notion of free will is challenged and the individual, for different reasons, has litde or no control. The personal health behaviours and lifestyle discourse mirrors the dominant discourse of poverty, suggesting that most health conditions are an individual responsibility and a consequence of poor decision-making, while good health is a matter of individual choice and control. A hegemonic middle-class perspective is evident in both dominant discourses of poverty and health, and in some cases one is adopted to support the other. Dominant poverty and health discourses present the controlled and healthy body as a symbol of moral rectitude and class association (White et al, 1995). "The ethic of health is often like the ethic of work... the Protestant world view extends to the body" (Crawford, 1984, p.67). The negative connotations of "welfare dependency" are supported by the biomedical and behavioural health discourses that 17 associate dependency with pathology and individual actions. Likewise, in 1987 the American Psychiatric Association newly coined "dependent personality disorder," a disorder most common in women, as an official psychopathology in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R). The stigma of dependency has escalated to such heights that both the biomedical and behavioural health discourses are used to enshrine dependency as a moral and psychological personality disorder, while undermining the possibility that dependency is a social relation of subordination or that interdependence is socially necessary (Wendell, 1996). Indeed power and domination tend to disappear (Fraser and Gordon, 1997). Understanding Poverty, Inequity, and Women's Health In order to explore the relationship between poverty and women's health, a social determinants of health perspective is most appropriate. This perspective acknowledges that individual change can rarely be achieved without harnessing social and environmental supports at a community level. The social determinants of health, however, have been interpreted and implemented in a variety of ways. Within this perspective, an individualistic approach often prevails — the individual is called upon to address the negative health effects of her oppression. For instance, in a recent study of 199 Canadian health care initiatives, the effects of poverty were addressed three times more often than were poverty's root causes (Williamson and Green, 1999). Despite problems with the interpretation and implementation of a social determinants of health perspective, it remains the most fitting perspective for understanding and addressing poor women's health. The experiences of being stereotyped, excluded, and made invisible are closely connected with social, cultural, and economic circumstances. If a woman does not fit within dominant middle-class discourses because she is a single mother who lives below the poverty line, she is stereotyped, marked out as the Other, and excluded from society. Thus understanding the dominant discourses of poverty and health are important for grasping 18 hegemonic conceptions of poverty and health, how people make sense of their experiences, and how some people are excluded, stereotyped, and invisible because they do not fit dominant criteria for belonging. Significandy, health equity researchers have advanced understandings of the relationship between poverty and health and have proposed ways to address health that do not rely on an individualistic discourse. Although the health equity field's conceptualizations of the gradient of health are useful, women's diversity and experiences remain inadequately conceptualized and measured. In this section I explore the major arguments forwarded by the health equity field and how these arguments both advance understandings of women's health and are gender-blind. The Gradient of Health One of the most pervasive and enduring observations in public health is the "gradient of health." This gradient can be pictured as a line on a graph that remains consistent across sex, age groups, cultural groups, countries, and diseases. The gradient of health shows that people who have the lowest socioeconomic status (SES)4 experience the highest rates of mortality and morbidity. As people move up the socioeconomic gradient, their health improves relative to the gradient (Deaton, 2002; Reid, 2002).5 There have been widespread efforts directed at finding plausible explanations linking socioeconomic status and health outcomes. Hypothesized explanations for this relationship include artifacts of measurement, health-related social mobility, biological and genetic factors, health-related behaviours, access to and use of health care services, exposure to different environments at home and at work, and psychosocial factors (Adler et al, 1994). Some researchers have examined the possibility that health is "self-selecting," in that health is a major determinant of life chances. This implies that health "selects" people into different social strata and that ill health leads to lower position in the social hierarchy. Poor health selects people who 19 have a job with fewer opportunities for control, who are at risk of unemployment, who live in deprived neighborhoods, who have fewer social networks, and who eat worse food and indulge in addictive and sedentary behaviour (Marmot, 2000). This hypothesis suggests that people's health is mostly determined independentiy of the conditions in which they live, and that healthier people are more likely to move up the social pyramid while unhealthy people move down. Two large data sets that have been used to study this possibility found that poor health does affect social mobility but that the size of the effect is too small to account for very much of the overall health difference (Power et al, 1990 and Wadsworth, 1986; cited in Wilkinson, 1996). Most evidence suggests that the economic, social, and pohtical conditions under which people hve their lives are major factors that determine whether they develop illness and disease (Raphael, 2001), and that exposure to unhealthy material circumstances is an indicator of relative income, or social position (Adler et al., 1994; Wilkinson, 2000).6 Wilkinson (1992, 1996, 2000) suggests that differences in life expectancies between countries can be explained in the first instance by differences in material wealth (GNP). After a certain threshold of material wealth and standard of living have been achieved there is a plateau effect such that differences in life expectancy and mortality rates can be explained by the relative differences in income between the wealthy and the poor. Per capita income is not as strongly related to life expectancy as is income distribution, with longer life expectancy associated with a greater proportion of income received by the least well-off 70% of the population. These effects of socioeconomic hierarchies are most strongly shown within countries rather than across countries, particularly in terms of life satisfaction. Indeed, the relationship between income distribution and health is well supported: alongside over 20 reports of an association between measures of income inequality and population health, there are only two reports (using the same data set) suggesting that the relation is weak or non-existent (Wilkinson, 2000). 20 The overall health of a society is more determined by the distribution of income among its members than the overall wealth of the society (Raphael, 2001). Since health is powerfully affected by social position and by the scale of social and economic differences among the population, what matters within societies is the effect of social relativities (Wilkinson, 1992, 1996, 2000; Raphael, 2001).7 The healthiest populations are those in societies that are prosperous and have an equitable distribution of wealth (Wilkinson, 1996; Health Canada, 1997). In the developed world, it is not the richest countries that have the best health, but the most egalitarian (Wilkinson, 1996). Psychosocial Health and Biological Expressions of Social Inequality Many health equity researchers argue that social position has its effect on health through psychosocial pathways, and have found that psychosocial conditions influence health and wellbeing (Wilkinson, 1996; Brunner and Marmot, 2000; Wilkinson, 2000; Raphael, 2001). Biological expressions of social inequahty are how people embody and biologically express experiences of economic and social inequality and produce social inequalities in health across a wide range of outcomes (Krieger, 2001). Brunner and Marmot (2000) developed a model that links social structure to health and disease through psychosocial, material, and behavioural pathways (Figure 1). Material circumstances are related to health direcuy and through the social and work environment. These shape psychological factors and health-related behaviours. The psychological and biological factors are the proximal causes of disease and tend to be the main focus of medical attention. These factors are the intermediates on the pathway from the social level to wellbeing or disease in individuals. In sum, social structure influences wellbeing and health, and the effects of social organization on population health are mediated by psychological and biological processes (Brunner and Marmot, 2000). 21 Figure 1 - The Social Determinants of Hea l th 8 Social structure Social environment Early life Genes Work Psychological Health behaviours Culture Material factors Brain Neuroendocrine and immune response Pathophysiological changes Organ impairment Wellbeing Mortality Morbidity Throughout this dissertation references to psychosocial health include stress, poor social networks, low self-esteem, depression, anxiety, insecurity, and the loss of a sense of control (Wilkinson, 1996). Although stress and depression,9 in particular, are distinct areas of research and have a number of theoretical and methodological particularities; unless differentiated they are included under the rubric "psychosocial health." According to Brunner and Marmot (2000), the power of psychosocial factors to affect health makes biological sense. The human body has evolved to respond automatically to emergencies. This stress response activates a cascade of stress hormones that affect the cardiovascular and immune systems. The rapid reaction of our hormones and nervous system prepares the individual to deal with a brief physical threat. The heart rate rises; blood is diverted to the 22 muscles; anxiety and alertness increase. This response is highly adaptive and may save life in the short term. But if the biological stress response is activated too often and for too long, there may be multiple health costs. These include depression, increased susceptibility to infection, diabetes, high blood pressure, and accumulation of cholesterol in blood vessel walls, with the attendant risks of heart attack and stroke (Brunner and Marmot, 2000).10 To emphasize psychological pathways does not mean that the basic cause of the problem is psychological or that it can be dealt with through psychological interventions. The point of cUstmguishing psychosocial pathways from exclusively material ones is to cUstinguish the social and economic problems affecting health through various forms of worry, stress, insecurity, from those that — like air or water pollution — affect health through material pathways, even if we are totally unaware of or unconcerned by them (Wilkinson, 1996). People's social and psychological circumstances can seriously damage their health in the long term. Chronic anxiety, insecurity, low self-esteem, social isolation, and lack of control appear to undermine mental and physical health (Brunner and Marmot, 2000). A sense of desperation, anger, bitterness, learned helplessness or aggression are all wholly understandable responses to various social, economic, and material difficulties. Often, prolonged stress from any of these sources is all it takes to damage health (Wilkinson, 1996). The Health Equity Field: Conceptual Limitations and Oversights The health equity field and theorizations of the gradient of health have made significant strides in conceptualizing the relationship between socioeconomic status and health. This field of research has given credibility to the social determinants of health perspective and has shifted health understandings from an exclusively biological and biomedical focus towards a more societal one. However, there remain limitations within this primarily epidemiological field, in terms of gender-blind conceptualizations and measurements of socioeconomic status, narrow 23 explanatory frameworks, a dearth of qualitative studies, and limited conceptualizations of women's health. Gender-Blind Conceptualizations and Measurements of Socioeconomic Status Gender is both present and absent in research on the socioeconomic gradient in health — present in the sense that including women in research studies is considered "standard good practice," but absent because litde systematic analysis of women's experiences or gender differences is being made (Laheima, Martikainen, Rahkonen, and Silventoinen, 1999). Research in the inequalities field makes general statements about the shape of socioeconomic gradients in mortality or morbidity, or about causal processes, without considering whether these apply differentially to men and women (Macintyre and Hunt, 1997). Women's experiences have been subsumed into a general "human" identity with the expectation of fitting traditional socioeconomic classification systems. When women have not fit these classifications, they have been treated as an aberrant population; when women have been added, observations on women are presented as additional results only, a procedure characterized as "add-women-and-stir" (Laheima et al., 1999). This is particularly surprising when psychosocial mechanisms are being discussed, since research in psychology, sociology, and anthropology suggest that women and men have different exposures to psychosocial threats and advantages, have different resources for dealing with threats, and may have different reference points for social comparisons (Macintyre and Hunt, 1997). There is an inherent male bias in measures of socioeconomic status because work is defined as primarily a male activity. Assumed differential roles and meanings of work for women and men (regarding the impact of work and work as a primary versus an additional role), the gendered segregation of work, and experiences of unemployment presuppose a relatively greater importance of work for men than for women.11 As well, the health equity field has typically 24 emphasized mortality data using differences between countries in terms of Gross National Product (GNP). The reliance on G N P exemplifies a gender bias since the G N P does not include the contribution of unpaid work in the home, almost all of which is done by women (Love, Jackson, Edwards, and Pederson, 1997). Additionally, whether or not women or Aboriginal people or disabled people were excluded from certain sectors of the labour market, whether or not they were paid less than white, able-bodied males for work of equal value, whether or not women and others were systematically excluded from positions of pohtical power, are often not criteria in comparisons of different countries (Love et al., 1997). Although we understand the nature of some of the gender differences in the socioeconomic gradient in health, the extent and complexity of these differences have not yet been fully documented. In most western societies, there have been ongoing conceptualization and measurement difficulties surrounding gender differences in socioeconomic status, especially when relying on the traditional classifications of education, occupation, and income. Women cannot be simply "added in" to existing analyses — a different analytic framework is required. Further empirical comparative research that sheds light on the different mechanisms behind women's poverty and on the different combination of beginning and ending events that mark periods of poverty is needed (Ruspini, 2001). Women's evolving roles and relationships both within and outside the home heighten these challenges. Measures of women's socioeconomic status must be reviewed and revised to accurately reflect the complexity and diversity of women's lives (Reid, 2002). Women's Poverty The poverty rate for women is higher than for men in every age group and disparity between socioeconomic groups is growing (Health Canada, 1997). Lone-parent families headed by women have the highest incidence of poverty for all family types, a situation that has improved 25 very little over the past decade, and older women are still more likely to be economically disadvantaged than their male counterparts. Currendy, almost 19% of adult women are living below the low income cutoffs13 - 41% of women over 65 and 56% of single mothers are low income (Health Canada, 1997).14 Poverty has been defined as a relative, multi-dimensional, and dynamic phenomenon.15 Poverty is also a gendered phenomenon — not only are more women than men likely to experience deprivation, but women's poverty is different from men's. A gender analysis of poverty is not so much about whether women suffer more from poverty than men (in numbers and/or intensity), but rather how gender differentiates the social processes leading to poverty. The trajectories leading to poverty or escape from destitution are gendered phenomena (Razavi, 1998). In North America the main causes of women's poverty are labour market inequities, domestic circumstances (marriage breakdown and motherhood), and welfare systems (National Council of Welfare, 2000; Ruspini, 2001). Recent research suggests that childbearing and society's definition of child rearing as a female activity place women and men in different structural positions with respect to the flow of information, resources in social networks, and financial compensation (Ballantyne, 1999). As well, women's poverty is highly associated with low education and the economic and social effects of racism and sexism, mcluding housing discrimination and exposure to domestic and community violence (Schroeder and Ward, 1998). Narrow Explanatory Frameworks The health equity field can be criticized for failing to adequately theorize the structural and ideological origins of social inequities, especially those based on gender, race, and class. Such oversights put health promotion at risk of failing to address the root causes of social gradients in health and unwittingly reproducing these inequities in the way health promotion is conceived and implemented (Poland, 1998). 26 The complex associations between social class and mortality are at the heart of the health equity debate. While suggesting a broadly focused discussion of the way social divisions — of gender, race, social class, and region — are reflected in the health experience of individuals, "health inequalities" have been more narrowly defined in terms of social class differences (Graham, 1990). As previously explained, researchers are searching for plausible "biological pathways" that might provide an explanation for social gradients in health. According to Poland (1998), this has resulted in the reduction of social processes to the status of statistical variables that belie a primarily bio-psychosocial and asocial perspective. Individual cognition and psychology are thus seen as the precursor to disease (Poland, 1998). Health is less related to people's absolute material living standards than to their position in society, as expressed by their income (Wilkinson, 2000). Income inequahty has been the focus of research on health and social life because it is an attractive measure for complex pohtical, economic, and social processes. Many societies, however, are built on overlapping patterns of oppression based on class, race, ethnicity, gender, and age, among other variables. Although health equity researchers have adopted a conceptual frame that equates these complex relationships with income (Capitman, 2001), it is problematic to rely on measures of socioeconomic status16 as a proxy for social class. Such measures inadequately reflect theorizations of social class and the complex lived realities of both women and men. Socioeconomic status is typically measured by a single variable or small sets of variables, thus failing to examine the whole range of the SES hierarchy and how SES may function most powerfully in combinations of variables. More complex, interrelated variables are needed (Adler et al., 1994). Current conceptualizations and measurements adopted by most health equity researchers limit the analysis, preclude different or alternative findings, and fail to consider the range of experiences living in poverty and how the composite of such experiences may influence 27 health.17 Dearth of Qualitative Studies Governments continue to rely almost exclusively on household surveys to generate easily quantifiable measures of poverty while case study material and other qualitative evidence are sidelined as anecdotal (Razavi, 1998). Surveys have been the primary data source mforming the vast majority of health equity research. However, there are inherent limitations in such data sets. They focus purely on household income levels while making it impossible to answer questions about intra-household distribution (Razavi, 1998). As well, surveys rely on literacy and English as a first language, and are incapable of capturing the experiences of living in poverty and how these experiences may differ between different groups of people. In order to better conceptualize and measure women's health in the context of women's lives, women's health researchers have called for more qualitative research methods. Qualitative methods are most congruent with a socio-environmental view of health and health promotion (McDonough and Walters, 2001) and may enable women's voices to be heard. Additionally, several authors have made a compelling case for the need for qualitative research on inequaHties in health as a corrective to the current bias in favour of positivist social epidemiology. Poland (1998) argues that research employing qualitative methods from a critical-interpretive paradigm is needed. Through conducting qualitative research one can enter into dialogue with women about not only their perceived needs but also the structural changes they believe will lead to improved health. Additionally, qualitative participatory approaches may increase participants' awareness of social constraints and can encourage participants to challenge their health barriers. Limited Conceptualizations and Measurements of Women's Health Epidemiological indexes of health status that include measures of morbidity and mortality are based on the notion of health as the absence of disease. The health equity field has largely 28 relied on measures of morbidity and mortality18 as a proxy for population health. These measures are most easily applied in large-scale epidemiological studies since they narrowly define health, are easily measured, and allow comparisons among populations. However, there are problems in relying solely on morbidity and mortality measures. They are imphcidy deemed important and representative of health irrespective of whether or not they exist alongside unjust and inequitable social relations based on gender, race, or other forms of exploitation. For instance, morbidity data do not indicate highly significant events in women's lives, such as rape, or other problems, such as arthritis (Redman, Hennrikus, Bowman, and Sanson-Fisher, 1988). The ends, such as low mortality rates, are defined as positive and desirable regardless of the context (Love et al., 1997). What is Women's Health? 1 9 Throughout life, the human experiences of birth, death, illness, and disability are embedded in social contexts. In any gender-dichotomized society, the fact that we are born biologically female or male means that our environments will be different, we will live different lives (Lorber, 1997), and we will have different experiences with health. Al l cultures characterize men and women as different types of beings, suitable for different kinds of tasks and entided to differing levels of economic, cultural and political resources (Doyal, 1995). These differences determine differential exposure to risk, access to the benefits of technology, information, resources and health care, and the realization of rights, all of which can influence health (World Health Organization, 1998). Indeed, women's everyday experiences must be understood within the context of the larger social organization and ideological structures generated from outside experience (Anderson, 1987). For decades mortality and morbidity rates have indicated that "women are sicker but men die quicker" — women live longer, men die prematurely, and women experience more morbidity 29 than do men. The life expectancy for Canadian women compared to Canadian men is almost seven years greater (Love et al., 1997). If one equates longevity with health, then women appear to be healthier than men (Reid, 2002). However, it is generally agreed that women suffer greater burdens of morbidity, distress, and disability (Doyal, 1995), including depression, stress overload, chronic conditions such as arthritis and osteoporosis, and injuries and death resulting from family violence. In general women present more acute medical problems, are hospitalized at higher rates than men, use more prescriptive medications, and report feeling less healthy (Greaves, 2000). Research over the last decade suggests that gender differences in health are rooted in social roles against the backdrop of some male biological disadvantage. Standard explanations for why women report more ill health have included biological or genetic risks, risks acquired through social roles and behaviours (the burdens of domestic and female role responsibilities, such as child rearing and housework), illness behaviour (women appear or act more sick), health reporting behaviour (women are seen to be more verbal or "complaining"), and differential health care access, treatment, and use (Hunt and Annandale, 1999; Macintyre, Ford, and Hunt, 1999; Reid, 2002). Women's health researchers and advocates suggest that health promotion and disease prevention activities must focus not only on diseases that are more common, more prevalent or more serious among women, but also on priority health issues identified by women themselves, women's diversity, and the determinants of health (Cohen, 1998, p.187). Women's health involves women's emotional, social, cultural, spiritual and physical well-being, and it is determined by the social, political and economic context of women's lives as well as by biology. This broad definition recognizes the vahdity of women's life experiences and women's own beliefs about and experiences of health (Philhps, 1995; cited in McComas and Carswell, 1996). Women's health is perceived as a continuum that extends throughout the lifecycle and that is critically and intimately related to the conditions under which women Hve. Examinations of 30 women's health require a social model of health that puts women's health needs at the centre of the analysis and focuses attention on the diversity of women's health needs over the lifecycle (Ruzek, Clarke, and Olesen, 1997).20 The traditional oppression and disempowerment of women must also be addressed at both personal and societal levels, thus broadening the approach. "Every woman should be provided with the opportunity to achieve, sustain and maintain health as defined by that woman herself to her full potential" (Ontario Women's Health Interschool Curriculum Committee; cited in Cohen, 1998, p.188). Poor and Unhealthy Women as "Other": A Theory of Exclusion Health equity researchers have advanced understandings of the relationship between poverty and health, and women's health researchers and advocates have built up a large body of work demonstrating the gender-blind theorizations of the health equity field with evidence of the intimate relationship between gender inequahties and health (Doyal, 1995; Stein, 1997; Annandale and Hunt, 2000; Doyal, 2000). Given the increasing rates and severity of women's poverty, the integration of these advances can further understandings of the social, economic, cultural, and pohtical practices and institutions that are implicated in the creation and perpetuation of women's poverty (Razavi, 1998). The complex ways that societal forces affect women's opportunities for health and experiences of health and wellbeing must be explored (Ruzek and Hill, 1986). Poor and unhealthy21 women are stigmatized and find little place in the dominant social fabric. According to Iris Young (1990), cultural imperialism is the construction of dominant societal meanings that render the perspectives of marginalized groups invisible while simultaneously stereotyping marginalized groups and marking them as Other. In so doing, the dominant group's experience and culture are universalized and established as the norm. "Other" groups are brought under the measure of dominant norms and are constructed as non-adherent, 31 deviant, and inferior. Since only the dominant group's cultural expressions receive wide chssernination, they become the normal, the universal, and thereby the unremarkable. "Others" are thus on the margins or excluded altogether from the social fabric, and are forced outside the definition of full humanity and citizenship (Young, 1990). This process of othering a person and labeling them as deviant depersonalizes the offending individual, who is then treated as a mere instance of a discreditable category, rather than as a full human being. The imposition of a stigma then provides a basis for collective discrimination against them (Schur, 1980). The selective stigmatization of Others is rationalized through the dominant means of interpretation and communication in society (Fraser, 1987). Typically, societal means of interpretation and communication are expressed through hegemonic discourses that fuel self-evident descriptions of social reality that normally go without saying and escape critical scrutiny (Fraser and Gordon, 1997). Particular interpretations of social life are enshrined as authoritative while others are de-legitimized and obscured, to the advantage of dominant groups in society and to the disadvantage of subordinate ones (Fraser and Gordon, 1997). Exclusion processes preserve poor women's multidimensional disadvantage and constructions as Other. Exclusion involves disintegration from common cultural processes, lack of participation in societal activities, alienation from decision-making and civic participation, and barriers to employment and material resources (Raphael, 2001). The explication of cultural, institutional, and material exclusion elucidates how the processes that create and perpetuate poor women as the Other influence their health. Poor women are culturally excluded as a consequence of being stigmatized, stereotyped, and invisible.22 Wendell (1996) asserts that we project rejected aspects of ourselves onto groups of people who are designated the Other. "We see 'the Other' primarily as symbolic of something else — usually something we reject and fear and project onto them" (Wendell, 1996, p.60). 32 Dominant cultural23 understandings portray women on welfare as draining resources and undermining social coherence. Women on welfare are seen to create division and difference — they are Others haunting the "norm" (Batsleer and Humphries, 2000), and are viewed as non-adherents of the individualized ideals of independence and self-control. They are culturally branded with having insufficient resolve, knowledge, and will-power to conform. In contrast, members of the powerful mainstream are portrayed as possessing the will-power and resolve to be independent, healthy, and working for pay. Cultural exclusion processes portray poor and unhealthy women in particular ways — as either in control of or at the mercy of poverty and ill-health, or, in other words, in control of or at the mercy of their "free will." In the notion of free will the origins and conditions of responsible behaviour, whether humans are free in what they do, or whether they are determined by external events beyond their control, are questioned (Audi, 1995). Thus, poor and unhealthy women, because they are poor and unhealthy, are othered because they have insufficiendy harnessed their free will. Dominant discourses rarely examine social, political, or economic factors; rather, the poor and unhealthy are told to secure their free will, take control of their situations, and make better choices and decisions. The processes of cultural exclusion do not realistically portray poor women's experiences — rather such processes are generally controlled by "right-thinkers" for whom being indigent or placing demands on the state to take care of the body are signs of moral weakness (Ingham, 1985). Consequently, only partial accounts of the world that are constructed from within particular historical contexts and that serve particular economic, pohtical, and social interests are presented (Kelly, 1996). Institutional exclusion is fueled by the deeply held assumption that moral agency and full citizenship require that a person be autonomous and independent. Poor women on government assistance depend on bureaucratic institutions for support and services. Being a dependent in 33 our society implies being legitimately subjected to the often arbitrary and invasive authority of social service providers and other public and private administrators, who enforce rules with which dependents must comply. Consequendy poor women are subjected to patronizing, punitive, demeaning, and arbitrary treatment by the pohcies and people associated with welfare bureaucracies. Paternalistic authorities24 construct the needs of people in their service, claiming to know what is good for them. Dependency in our society implies, as it has in all liberal societies, a sufficient warrant to suspend basic rights to privacy, respect, and individual choice (Young, 1990). While those who are more socially included have greater access to economic, educational, and social resources and support (Shaw et al., 2000), cultural and institutional exclusion processes rationalize poor women's material exclusion. Exclusionary processes confine poor women's material lives, mcluding the resources they have access to and the opportunities they have or do not have to develop and exercise their capacities (Young, 1990). Poor women are economically excluded since they are expelled or retreat from equal participation in economic and social life (Young, 1990). Metaphorically, exclusion "trickles down." It trickles from cultural assumptions and stereotypes, fuels punitive institutional policies and practices, and legitimizes material deprivation. Exclusion as a "trickle down" means that linear measures and conceptualizations of health and poverty are not enough. What matters are women's multi-faceted experiences living in poverty, how these experiences are socially isolating, and the effects on health. Cultural exclusion processes, such as stereotyping, surveillance, and threatening, shame poor women. Institutional exclusion trickles from cultural exclusion and denies people the opportunity to participate in and contribute to society and to enjoy an acceptable supply of goods and services. It functions through discriminatory practices that create barriers to accessing services and 34 resources. Meanwhile, poor women's material exclusion and scarcity are legitimized. Cultural, institutional, and material exclusion lead to human misery, loss of potential, and poor health (Shaw et al., 2000; Raphael, 2001b). Unpacking the "Other": Provisional Identities and Women's Diversity In a society that systematically excludes and devalues poor and unhealthy women, only partial and distorted understandings of ourselves and the world around us can be produced (Harding, 1987). In many instances the weight of ideology has systematically distorted people's views of their world and their own capabilities (Rahman, 1983) and the result can be shame, passivity, and a resignation to the status quo as an unchangeable and natural experience (Comstock and Fox, 1993). Some theorists believe that shame is the key to social conformity, forcing the shamed to avoid embarrassment and being thought different, inadequate, or stupid (Wilkinson, 2000). "Unacknowledged shame plays a central role in causing subjects to yield to group influence, even when it contradicts their own direct perceptions" (Scheff, Retzinger, and Ryan, 1989, p. 184). Although stereotyped and inferior images may be internalized by group members, "deviants" must also make various efforts to avoid or counteract the social and psychological impact of the stigma (Goffman, 1963). Despite the power of dominant exclusion processes, oppressed people often refuse to adhere with the dominant group's devaluing, objectifying, stereotyping visions. "Individuals make use of the discursive resources available to them while at the same time positioning themselves in ways that represent the least risk in terms of challenges to existing systems of knowledge and belief' (Croghan and Miell, 1998, p.449). Although stereotypes and shame can produce passivity, they can also facilitate a kind of agency to re-create identity (Sedgewick, 1993). According to hooks (1990), shame and marginalization can give birth to a process of poUticization, become a place of action, and enable a reclamation and reinterpretation 35 of socially derided identities (hooks, 1990). Though Others are seen as non-conformists and deviants, they may reclaim and reinterpret their socially derided identities. Identity is the social self that is named and experienced. It is a contingent and variable sense of self, both conscious and unconscious (Ristock and Pennell, 1996). Individuals inhabit different identities that may be stressed or muted depending on the situation (West and Fenstermaker, 1995). According to Fraser (1997), people's social identities are complexes of meanings and networks of interpretation. To have a social identity is to live under a set of descriptions that is drawn from the fund of interpretive possibilities available to individuals in specific societies. Since everyone acts in multiple social contexts, the different descriptions comprising any individual's social identity fade in and out of focus. People's social identities are fashioned and altered over time. Under conditions of inequahty, social groups are formed and at times work towards emancipatory social change (Fraser, 1997). Feminists grapple with the substantial theoretical challenge of how to honour and appreciate diversity, while also recognizing how difference is constructed (Hunt and Annandale, 1999). Postmodern feminists reject master narratives and suggest mini-narratives that are provisional, contingent, temporary, and relative, and that provide a basis for the actions of specific groups in particular circumstances. Feminist postmodernists stress fragmentation - of language, time, the human subject, and society itself (Batsleer and Humphries, 2000). As such, any naming or categorizing tends to call attention to similarities and to neglect differences (Acker, Barry, and Esseveld, 1991; Martin, 1994). Paradoxically, postmodern feminist attempts to unmask the differences among women have imposed false unity on their research. In overcompensating for a failure to acknowledge the differences of race, class, and ethnicity, feminists gave privileged status to a predetermined set of analytic categories and affirmed the existence of nothing but difference. In other words, in trying to avoid the pitfall of false unity, they walked straight into 36 the trap of false difference (Martin, 1994). As feminists we must expose both the links and contradictions between ourselves and other citizens of the world. We must resist the temptation to mask crucial differences with the notion of universality and to deny generalities for fear of essentialism. Consequentiy, feminist postmodernists encourage the development of multiple discourses that are responsive to the diversity of women's experiences (Ristock and Pennell, 1996). The provisionality of identities is accepted as well as their daily solidity in the relations of power. Identities are useful and meaningful; in both pohtical organizing and personal lives, identities are adopted because they make sense at that moment, though they are not definitive or everlasting (Spivak, 1990). Identity is saturated with the voices of others and needs to be understood not as a unified and static entity but as constructed in order to achieve specific strategic and presentational ends (Croghan and Miell, 1998). People come into interactions by assuming situational identities that enhance their own self-conceptions or serve their own needs, which may be context specific rather than socially or culturally normative (Angrosino and Mays de Perez, 2000). What people say in interviews, in writing, or in their everyday interactions can differ from what they really think, and attitudes and behaviour may not always match each other. Language is by nature metaphorical, figurative, and context-dependent and not very successful at mirroring complex circumstances (Alvesson and Skoldberg, 2000). Language can also be used to construct individual subjectivity in ways that are historically and locally specific. Reality is continually constructed and reconstructed moment by moment through linguistic and social practices and self-conscious reflection. This multiple and changing world includes multiple and changing selves. Persons are understood to not possess a single, stable, internal identity but to have many selves which are constructed and then reconstructed depending on the context (Drew and Dobson, 1999). 37 In my efforts to study poor women's health, it was important to understand the cultural, institutional, and material dimensions of exclusion, how the dominant discourses of poverty and health construct poor and unhealthy women as the Other, and the ways that these multifaceted experiences of exclusion influenced women's health. Feminist Action Research: A Response to Poor Women's Exclusion Although health has long served as a barometer of the economic and social conditions under which people live (Walters, Lenton, and Mckeary, 1995), the development of appropriate tools to measure the impact of gender on women's health has lagged behind conceptualizations about social and structural influences (Reid, 2002). Oversights in the health equity field include the dearth of qualitative studies and the narrow gender-blind conceptualizations of women's health and poverty. Additionally, hegemonic discourses of poverty and health that favour middle-class values and experiences necessitate research efforts that explore poor women's experiences. Feminist action research (FAR) is a conceptual and methodological framework that attempts to address these needs and limitations. Feminist action researchers typically use qualitative research methods to generate in-depth understandings of women's experiences and put women's diversity at the centre of the analysis. FAR strategies attempt to be inclusive, participatory, collaborative, and to elucidate poor women's exclusion. Indeed, FAR can be seen as a research tool to address poor women's complex experiences — in terms of better understanding the factors that perpetuate their exclusion, appreciating the diverse and often disparate ways that poor women negotiate their lives, and responding to social injustices through advocating collective action and social change. Feminist action research is a blending of participatory research and feminist theory. In this section I describe participatory research; the genesis, goals, and principles of FAR; and feminist action researchers' current struggles with academic co-optation and marginalization. Although 38 these challenges have not been fully rectified, I saw FAR as an appropriate and useful framework for examining the three research questions that guided this dissertation. Participatory Research Participatory research (PR) arose during an era of reflection and self-questioning witliin social research.25 At issue were the purposes of research, definitions of objectivity, power relationships between the researcher and the researched, ownership of research results, and the ethics of data collection and reporting (Maguire, 1987; Fals-Borda, 1991; Green, George, Daniel, Frankish, Herbert, Bowie, and O'Neill, 1995). In response to these issues PR is value-driven and self-consciously not value-neutral. PR integrates subjectivity into a scientific analysis and to the work of structural and social transformation (Vio Grossi, 1981). Participatory researchers advocate methods of data collection and analysis that are grounded in the context of the community and contribute something of value to the community in which the research is conducted (Herbert, 1996). Most generally, participatory research is a process that combines three activities: research, education, and action (Hall, 1981). PR involves a social action process that is biased in favor of dominated, exploited, poor, or otherwise ignored women, men and groups. Participatory researchers work "with" rather than "for" the researched, breaking down the distinction between researchers and the researched while legitimizing the knowledge people are capable of producing (Fals-Borda, 1991). Consequendy, participatory researchers outline and utilize explicit processes to facilitate ordinary people's reflection on and analysis of their reality. They attempt to involve participants in the entire research process, mcluding an action phase, and present people as researchers in pursuit of answers to questions of daily struggle and survival (Tandon, 1988; Fals-Borda, 1991; Nyden and Wiewel, 1992). The people studied make decisions about the study format and data analysis. This model is designed to create social and individual change by altering relationships in the 39 project (Reinharz, 1992). Participatory researchers see no contradiction between goals of collective empowerment and the deepening of social knowledge (Hall, 1992), and attempt to adhere to the tenets of community empowerment, emphasizing increased power and control across multiple sites of practice. PR involves a critical problem-solving process of diagnosing, action planning, action taking, evaluating, and specifying learning (Brown and Tandon, 1983; Hall, 1992, 1993; Park, 1993; Freire, 1996). Participatory researchers espouse that through a dialectical process of collective reflection and action — "conscientization" — the community and its constituent organizations and individuals foster a sense of identification and shared fate. Through developing the skills and resources to engage in the cyclical process of diagnosing and analyzing problems, they plan, implement, and evaluate strategies aimed at meeting identified needs (Israel, Checkoway, Schulz, and Zimmerman, 1994). Community members, organizations, and the community as a whole gain increased influence and control, that is in turn associated with improved health and quality of life. Together participatory researchers and community members strive to maximize both increased knowledge and understanding of a given phenomenon and take action to change the situation (Israel et al., 1994). Action and evaluation proceed simultaneously in a fluid approach that is constantiy evaluated, rather than a traditional research project where the design is thought out at the begmning, implemented in a carefully controlled way, and finally evaluated (Reinharz, 1992). Thus the credibihty-vahdity of PR knowledge is measured according to whether actions that arise from it solve problems and increase participants' control over their own situation (Greenwood and Levin, 1998). Feminist Action Research Participatory research taught me the necessity of being explicit about personal choices and values in the research process. Feminism taught me to recognize that the personal is pohtical (Maguire, 1987, p. 5). 40 Despite the ideals advanced by participatory researchers, gender and women's diversity were made invisible by early assumptions that women could be automatically included in terms such as "people," "community," or "the oppressed" (Hall, 1993). In traditional research as well as in participatory research, women have been largely excluded from producing dominant forms of knowledge. Maguire (1987) has pointed out the distinct silence around gender and women in PR discourse, calling it the "androcentric filter." Participatory research is built on a critique of positivism that often ignores and repeats many of the androcentric aspects of dominant social science research. Freire (1996) maintained that domination was the major theme of our epoch, but his conscientization tools ignored men's domination over women. Freire, like others "does not depart from taking androcentricity as the norm, and consequendy, feminists need to do the work for women that he did for men" (Klein, 1983). Women's ways of seeing were not mentioned until 1981 and the general discourse of women has been excluded (Maguire, 1987). Without recognition of, and attention to, its male biases, participatory research cannot be truly emancipatory for all people (Hall, 1992). Feminist researchers have attempted to address the androcentric biases inherent in traditional and participatory research, yet have not articulated distinct methods or methodologies. Within "feminisms"2 6 there is no single or monolithic method, methodology, or theoretical base of feminist scholarship; in fact, there are competing theoretical foundations and varied methodologies (Maguire, 2001). As well feminists have not agreed upon one definition of research (Maguire, 1987). There has been resistance to a rigid, dogmatic "correct" feminist methodology because it may reinforce domination and limit knowledge. However, there is a great danger in "feminisms" becoming co-opted by mainline methods that reinforce unequal power relations unless we commit ourselves to a distinctive methodology (Cancian, 1992). Some common features of feminist methodology include focusing on gender and inequahty 41 and using qualitative methods to analyze women's experience. However, very few studies adopt the more radical methods of including an action component, using strong participatory methods that give participants substantial control over involvement in the study, and critiquing the power relations in academia (Cancian, 1992). Despite embracing a call for transformational structural and personal action (Mies, 1991; Maguire, 2001), feminists have not outlined a clear strategy for elirnmating androcentricism from research, nor have they "given adequate attention to the envisioning of truly emancipatory knowledge-seeking" (Harding, 1986, p. 19). Feminist theories, epistemologies, and methodologies have inspired and grounded many action researchers' work (Maguire, 2001), yet there are only a handful of systematic attempts to link feminism and action research (Greenwood and Levin, 1998). Aspects of participatory research and feminist research cohere ontologically and epistemologically as both seek to shift the centre from which knowledge is generated (Hall, 1981). As well, they share an avowed intent to work for social justice and democratization (Lather, 1991; Greenwood and Levin, 1998; Maguire, 2001). By combining feminist research's critique of androcentricism with participatory research's emphasis on participation and social change, feminist action research provides a powerful approach to knowledge creation for social and personal transformation (Maguire, 1987). Despite the fact that few participatory or feminist researchers use "feminist action research" nomenclature, FAR guided my research and underscored my research values and goals. I define feminist action research as a conceptual and methodological framework that enables a critical understanding of women's multiple perspectives and works towards inclusion, participation, action, and social change while confronting the underlying assumptions the researcher brings into the research process. Feminist action researchers facilitate building knowledge to change the conditions of women's lives, both individually and collectively, while 42 reconstructing conceptions of power so that power can be used in a responsible manner (Ristock and Pennell, 1996). Feminist action research is a tool, not a panacea, that involves a particular way of looking at the world and thinking about research participants (Maguire, 1987). Since many poor women are excluded from social processes and their communities, open and flexible theory-bunding grounded in a body of empirical work ceaselessly confronted with, and respectful of, women's day to day experiences is needed (Lather, 1991). Praxis: Developing A Feminist Action Research Framework Participatory research and contemporary feminism share a number of major underlying features centering on the analysis of pohtical economy and praxis (Maguire, 1993). The notion of praxis challenges the theory-practice relationship and raises problems associated with value-free science. The term "praxis" originates from the Greek word prasso meaning "doing" and "acting", in contrast with the theoretical designs of epistemology, theoria 2 7 (Audi, 1995). Lai (1996) refers to pedagogical praxis — the erasing of the boundaries between theory, methodology, and practice, and between field and home (Lai, 1996). Other researchers define praxis as a dialectical process of collective reflection and action (Kirby and McKenna, 1989; Israel et al., 1994); the joining of theory and action so that each is informed by and changes through its relation with the other (Ristock and Pennell, 1996); and the integration of knowing and doing (Kirby and McKenna, 1989). For praxis to be possible, theory must not only illuminate the lived experience of social groups, it must also be iUuminated by their struggles (Lather, 1991). Theory is thought of as experience-based (Park, 2001), and reflection itself is embedded in praxis, not separate from it. Action upon reality and analyses of that learning may change awareness of the nature of problems and the sources of oppression (Gaventa and Cornwall, 2001). Indeed the processes of uncovering marginalized knowledges, giving priority to these knowledges, and enabling the 43 collaborative development of action strategies are simultaneously theoretical and methodological. Woven widiin and between each of the following characteristics of a FAR framework is the notion of praxis — theory, methods, and practice in dynamic interplay. A Feminist Action Research Framework For the purposes of this dissertation I have developed a feminist action research framework based on the principles of inclusion, participation, action, social change, and researcher reflexivity. In this section I present these principles as ideals, and recognize that they do not always reflect the reality of engaging in FAR. In chapter 7 I examine the realities of these FAR ideals and the ways that they may be reframed to more realistically portray feminist action research. Inclusion The first principle of feminist action research is inclusion. To include is to cause to be a part of something, to consider with, or to put into a group (Morris, 1982). Feminist action researchers contend that no social practices or activities should be excluded as improper subjects for public discussion, expression, or collective choice; no one should be forced into privacy (Young, 1990). Gender and women's daily experiences are central in the construction of theoretical frameworks and feminist action research methods. Women's experiences are central in several ways - in understanding patriarchy as a system of domination and oppression, in identifying key issues and questions in all phases of the research process mcluding action and evaluation, and in giving explicit attention to how women and men as groups benefit from the project.28 Feminist action researchers expose the inadequacy of androcentric research and its partial, inaccurate, and incomplete representation of human experience when women are muted (Maguire, 2001). By begmning with the experiences and research needs of women who have 44 traditionally been silenced, the process of knowledge production is transformed and the ideological power base is challenged (Kirby and McKenna, 1989). Feminist action researchers seek to connect the articulated, contextualized personal with the often hidden or invisible structural and social mstitutions that define and shape our lives. Feminist action researchers attempt to make possible the development of strategies and programs based on real life experience rather than theories or assumptions, providing an analysis of issues based on a description of how women actually experience those issues (Barnsley and Ellis, 1992). Drawing on W O A W as an example, diverse women on low income were included; fundamentally, their experiences, barriers, and needs were seen as central to WOAW's vision, mandate, and research processes. Participation Inclusion is a precursor to the second principle of feminist action research - participation. To participate is "to take part, join, or share with others" (Morris, 1982). Feminist action researchers are committed to making women's voices more audible and faciktating women's empowerment through "ordinary talk" (Maguire, 2001). Some researchers have attempted to outline different kinds and levels of participation. Herbert (1996) discussed levels of participation as the "seven Cs," ranging from collusion, co-opting, and coercion, convincing, coordination, and cooperation, to true collaboration. It is essential that researchers in partoership with communities be clear about the level of participation they are expecting and inviting (Herbert, 1996). It is also important to question who is participating, why they are participating, to what degree and in what phase of the project they are participating, and where the true power lies (Wallerstein, 1999). Feminist action researchers contend that at all stages of research, full collaboration and participation should occur between the researched and the researcher and among the research 45 participants. Shared decision-making must be promoted. Collaboration is seen as central to budding knowledge and co-creating meanings that will engage the researcher and the research participants in mutual dialogue (Van Den Bergh, 1995). Participation is a value that guides W O A W organizational processes, particularly sharing stories and experiences, hearing all voices, learning from one another, honouring different levels of involvement, and fostering different ways to be involved in meetings, activities, and research. Individual and Collective Action The third principle of feminist action research is individual and collective action. Action is defined as "the transmission of energy, force, or influence" (Morris, 1982, p.13). Action is a dynamic process (McWiUiam, 1996); what action looks like is based on one's social, economic, and pohtical situations; and action can occur on both individual and collective levels. People with problems figure out what to do by first finding out their causes and then acting on insight (Park, 2001). Through action we learn how the world works, what we can do, and who we are -we learn with mind and heart — and this is how we become aware and emancipated, or how we learn our powerlessness. Action is an integral part of reflective knowledge, and can be conceptualized as speaking, or attempting to speak, to validate oneself and one's experiences and understandings in and of the world (Gordon, 2001). Lorde (1984) advocates turning silence into language and action. Action can produce changes in participants that go beyond intellectual understandings. Thus the processes of individual and collective action can encompass a wide range of endeavours — leaving the house, managing one's day-to-day life and health issues, attempts at negotiating the welfare and health care system, the development of an identification and shared fate with a group, the growth of a sense of self and collective efficacy, the belief that action is possible, and the capacity to develop individual and collective strategies for action (Israel et al., 1994). The 46 W O A W organization provides useful examples of individual and collective actions. Leaving the house, going for a walk, learning to manage depression, and joining a support group are examples of individual actions taken by W O A W women. Collective actions occur when groups of W O A W members, either formal or informal, initiate an activity, meeting, community involvement, or pohtical activism for themselves, for other W O A W women, or for women in their communities. Social Change The purpose of feminist action research is to create new relationships, better laws and improved institutions (Reinharz, 1992). Social change can be envisioned as the outcome of deliberate individual and collective actions. Many feminists envision social change as moving towards "a society that would develop individuality but shift the balance from individual rights towards the rights of the majority and the collective, and that would validate the pursuit of the common good rather than individual self-interest" (Adamson, Briskin, and McPhail, 1988, p.101). Social change is the process of altering the initial situation of a group, organization, or community in the direction of a more liberated state (Greenwood and Levin, 1998). Maguire (2001) envisions social change as "the long haul struggle to create a world in which the full range of human characteristics, resources, experiences, and dreams are available to all our children" (p.66). It is important to note that although most feminist action researchers avow social change as the ultimate goal of their research endeavours, it remains poorly understood and conceptualized. From the FAR hterature it is difficult to know exacdy what it looks like and what can count as social change. Not surprisingly, the ambiguity of social change also posed problems for the W O A W organization. The W O A W organization has a vision of social change that included all women in the community being empowered, respected, and treated with dignity. To be sure, 47 accompHshing WOAW's vision requires a number of individual and collective actions. Although social change is an implicit goal shared by most W O A W members, few of them articulated specific strategies to realize it. Consequentiy, in this dissertation's research questions and analysis social change is framed as "doing the work of social change" and "taking action towards social change." Researcher Reflexivity The final principle of feminist action research is researcher reflexivity. Feminist action researchers advocate high standards of reflexivity, openness, and transparency about the choices made throughout any empirical study (Kirby and McKenna, 1989; Hertz, 1996; Reay, 1996; Rose, 1997; Tom, 1997; Edwards and Ribbens, 1998; Coleman and Rippin, 2000). Reflexivity is the capacity to locate one's research activity in the same social world as the phenomena being studied and to explain the nature of research within the same framework as is used to theorize about the objects of study (Reason, 1994). Feminist action researchers need to openly and honesdy recognize their "conceptual baggage" (Kirby and McKenna, 1989), and consider the implications of their power. Feminist action researchers are expected to be transparent and to appreciate methodological, epistemological, and pohtical influences, contradictions, and complexities in all stages of research (Ristock and Pennell, 1996). In order to be cognizant of the class and power differences between myself and the research participants, in chapter 3 I describe my strategies for being deliberately reflexive. The Marginalization and Co-optation of Feminist Action Research Ideals Feminist action researchers are critical of traditional authoritarian approaches to knowledge generation that use only a tiny fraction of the knowledge generated to confront important problems (Greenwood and Levin, 1998). In most societies the gap between those who have power over the process of formally recognized knowledge generation and those who do not has 48 reached dimensions no less formidable than the gap in access to the means of physical production (Rahman and Fals-Borda, 1991). Although feminist action researchers may use methods that intend to "give the research subjects more power," it is not clear that they have succeeded (Cancian, 1992, p.627). Academic feminists have tended to maintain control over research projects and knowledge creation as have conventional non-feminist researchers, rarely empowering the women they study (Maguire, 1987). By mamtaining this control and distance, most feminist scholarship ends up furthering the gap between the researcher and the researched and benefiting the researcher more than those studied. This behaviour undercuts some of the goals set forth by feminist action researchers and reproduces aspects of traditional academic research. Although feminist action researchers demand that the researcher give up some of these controls and share them with others, this is still rare in academic feminist academic research. Feminist fieldworker-scholars have tended not to take up the call of more participatory research and have held on to the reins of research and writing (Wolf, 1996). At times the esoteric language of much feminist scholarship keeps feminist action researchers from putting their research to use. Esotericism effectively keeps the knowledge constructed by feminist scholars out of the hands of feminist activists and practitioners. It can also keep feminists inside the academy from understanding one another (Martin, 1996). Research with a more participatory component, like feminist action research, challenges feminist scholars to practice what they believe and may preach - more egalitarian approaches to empowerment that are with, and not simply for, the researched population. While this is deeply connected to the structure of power and privilege in the academy, it nonetheless points to a highly problematic contradiction among feminist scholars (Wolf, 1996). Such difficulties arise in part because there are different uses of knowledge in the academy 49 from those in community or workplace situations. Knowledge is a commodity by which academics do far more than exchange ideas — it is the very means of exchange for the academic pohtical economy (Hall, 1992). Academics are under economic, job survival, or advancement pressures to produce in university-approved ways. The dominant methodology of an academic discipline usually supports the existing power structure of the chscipline and its environment. Often, leaders in academia favour complex methods that outsiders cannot use or understand and devalue the knowledge of lower-status colleagues and the general pubhc (Cancian, 1992). Pubhc confidence in scholarly objectivity has consequentiy been eroded by growing awareness of the material interests and ideological prejudices that have a profound influence on the consciousness and practices of researchers and on the conclusions drawn from their work (Ristock and Pennell, 1996). Indeed some argue that "the word 'academic' is a synonym for irrelevant" (Alinksy, 1969, p.xi; cited in Nyden and Wiewel, 1992). In many cases the academic reward system renders traditional cuscipline-bound research the main form of legitimate research in the eyes of university tenure and promotion committees (Nyden and Wiewel, 1992). People who challenge this methodology risk being marginalized, rejected, or otherwise punished (Cancian, 1992). These structural pressures impede academic engagement in participatory and feminist research processes (Hall, 1992). FAR discourse promotes shifting power or structural changes through the knowledge generated. Such radical methods direcdy challenge the system of inequahty in academia and may draw opposition (Cancian, 1992).29 Currentiy few academic researchers use PR or FAR processes (Hall, 1992) since they are time-consuming and chaUenging to academia's status quo. Paradoxically, however, some disciplines, including (but not exclusively) health promotion, education, women's studies, and organizational studies, promote the feminist action research ideals of inclusion, participation, 50 action, and social change. In fact, in these fields the proliferation and popularization of participatory and feminist research have diluted the meanings of these terms - they have become open, multiple, contested, and contestable, and have adopted the appearance of slogans or mantras that are so charged with multiple meanings that they are now empty and almost meaningless (Poland, 1998). For example, the popularization of PR and FAR has caused many activists to worry legitimately about co-optation of their perspectives for the purposes of obscuring and blunting democratic initiatives (Greenwood and Levin, 1998). In many cases participation has been exclusively looked at as a way to improve the final product rather than as a tool for developing a process of social transformation. With new words, but old techniques, the separation between the subject and the object of research persists and the dominant features of such practices continue. FAR and PR, devised for liberation, are "converted into a new, and perhaps more efficient, tool for manipulation" (Vio Grossi, 1981, p.44). Such problems are related to idealized characterizations and promises of FAR and to vague definitions (Vio Grossi, 1981). The challenges of FAR, as described by Vio Grossi (1981), propelled me to provide clear definitions of the FAR principles that guided this dissertation and to examine the lofty ideals and lived realities of engaging in FAR (chapter 7). As a graduate student researcher I had to consistendy examine my own role in and expectations of the research process, and to clearly articulate what it means, both theoretically and methodologically, to participate in feminist action research. In this chapter I argued that mainstream understandings of the gradient of health inadequately capture the complexity of poor women's experiences. Poor women, as the Other, are culturally, institutionally, and materially excluded from participating in society. They are stereotyped and invisible, and as a consequence forced to negotiate their socially derided 51 identities. In problematizing mainstream conceptual and methodological approaches in the health equity field, developing a theory of women's exclusion, and exploring the dominant discourses of poverty and health, I attempted to advance understandings of exclusion and poor women's health. Adopting a FAR framework can enable more complete understandings of the factors that perpetuate women's exclusion and poor health and the diverse and disparate ways that poor women negotiate their lives, while providing strategies for responding to social injustices through advocating collective action towards social change. Notes 1 B.C. Benefits includes social assistance, disability benefits, and seniors' pensions that are paid by the provincial government (British Columbia). "Employable" one-parent families with one child receive $845.58 per month on social assistance; one-parent families with two children receive $935.58. Single people on Disability Benefits I receive $802.92 per month; one-parent families with one child receive $896.58. Disability Benefits II recipients who are single adults receive $981.42, and recipients who are single parents with one child receive $1 050.08. For adults over the age of 65, a person living alone receives $786.42 per month; two adults living together over the age of 65 receive $1 329.06 per month (B.C. Ministry of Human Resources, 2002). Throughout this dissertation B.C. Benefits is also referred to as "assistance," "the system," and "welfare." 2 Although the "health and income equity" field originated primarily in Europe, North American "socioeconomic status and health" researchers also study the relationship between health and poverty. The majority of researchers in this field are social epidemiologists. 3 It should be noted that in Canada universal health care is increasingly contested as fiscally unrealistic and unsustainable. 4 Socioeconomic status is "a composite measure that typically incorporates economic status, measure by income; social status, measured by education; and work status; measure by occupation" (Adler, Boyce, Chesney, Cohen, Folkman, Kahn, and Syme, 1994, p. 15). Socioeconomic status refers to the position in the social hierarchy that gives individuals relative power and recognition due to wealth and certain forms of income (Ballantyne, 1999). 5 A series of studies in the United Kingdom document how those living on lower incomes are more likely to suffer and die from cardiovascular disease — and a number of other diseases — at every age. A recent study found significant differences in overall death rates between those in the lowest two income groups and those in the highest two income groups in England and Wales. Lower income women had a 55% greater chance of dying than those with higher incomes. In the United States, low-income Americans have a higher incidence of a range of illnesses including cardiovascular disease. In Canada, data on individuals' income and social status are not routinely collected at death, so national examination of the relationship between income and death from various diseases must use census tract of residence to estimate individuals' income. In both 1986 and 1996, those Canadians living within the poorest 20% of urban neighborhoods were much more likely to die from cardiovascular disease, cancer, diabetes, and respiratory diseases — among other diseases — than other Canadians (Raphael, 2001). Latest estimates are that 23% of premature years of life lost prior to age 75 in Canada can be attributed to income differences. That is, 23% of all of the premature years of life lost to Canadians is accounted for by the differences that exist among wealthy, middle-income, and low-income Canadians (Raphael, 2001). According to a Health Canada report, Canadians with the lowest income were five times more likely than those from the highest income groups to report their health as only fair or poor, two times more likely to have a long-term activity limitation, and only one-third as likely to have dental insurance (Health Canada, 1999). 52 6 Relative poverty, or relative deprivation, is the degree of disadvantage, or, the proportion of the population who fall behind and the distance they fall behind. Relative poverty socially and materially excludes people from society and enforces a sense of being disadvantaged and stigmatized in relation to those better off (Wilkinson, 2000). Relative deprivation refers to the disadvantaged position of an individual, family, or group relative to the society to which they belong, and focuses on the condition of deprivation as well as the lack of resources (Shaw et al, 2000). People are relatively deprived if they cannot obtain the conditions of life - that is, the diets, amenities, standards, and services — that allow them to play the roles, participate in the relationships, and follow the customary behaviour which is expected of them by virtue of their membership in society (Raphael, 2001). Absolute poverty, or absolute deprivation, refers to living in material scarcity and not having the resources for living essentials such as food, shelter, and clothing. Absolute poverty is a lack of the means of providing material needs and comforts, deficiency in amount, insufficiency, paucity, and scantiness (Morris, 1982). 7 While there have been fluctuations in the prevalence of low income and of average income over the last 15 years, the trends in income disparity have been more consistent. In Canada the incomes of the wealthiest 10% of families with children grew 14% from 1981 to 1996, while the incomes of the poorest 10% declined 5% during this period (Health Canada, 1999). 8 Taken from Brunner, Eric and Michael Marmot. (2000). Social organization, stress, and health. Social Determinants of Health. Ed(s). M . Marmot and R. G. Wilkinson. Oxford, U K : Oxford University Press, pp. 17-43, by permission of Oxford University Press. 9 Stress, a term widely used in the biological, physical, and social sciences, is a construct whose meaning in health research is defined in relationship to stressful events, responses, and individual appraisals of situations. Common to these definitions is an interest in the process in which environmental demands tax or exceed the adaptive capacity of an organism, resulting in psychological or biological changes that may place the person at risk for disease (Cohen, Kessler, and Underwood, 1995). Stress has been used not only to refer to extreme environmental or psychosocial conditions, but also as a substitute for what otherwise might have been called anxiety, conflict, emotional distress, ego-threat, frustration, threat to security, tension, and arousal (Mikhail, 1981). Although stress research has been focused primarily in psychology, the study of stress is truly interdisciplinary. An environmental stress perspective focuses on environmental demands, stressors, or events, while a biological stress perspective focuses on activation of the physiological systems that are particularly responsive to physical and psychological demands (Cohen et al., 1995). Depression is a disabling condition characterized by feelings of sadness, sometimes accompanied by a sense of irritability, helplessness, and hopelessness (Health Canada, 1999). From a clinical perspective, depression is the most common primary care psychiatric diagnosis, with a lifetime prevalence of major depression at about 5% for men and 10% for women (Nicoloff and Schwenk, 1995). Women are twice as likely as men to be depressed and the duration of their depression is likely to be longer (Ballantyne, 1999; Health Canada, 1999). Recent surveys report that 20% of Canadians feel depressed some of the time, that younger adults experience depression more often, and that people with higher levels of education and income report less depression (Canadian Fitness and Lifestyle Research Institute, 1998). About 9% of people in the lowest income group were depressed, compared with 3% of Canadians in the highest income group. In addition, sad people in the lower middle income group were by far the most likely (23%) to be blue for more than half the year, while sad people in the highest income group were the least likely (6%) to be blue for the same length of time (Health Canada, 1999). 1 0 A recent study on the role that stress plays in disease identified the psychological and biological pathways through which exposure to adverse psychosocial circumstances — of which low income is one of the most potent — leads to the onset of cardiovascular disease. The social environment can create adverse conditions that produce the "fight or flight" reaction. This works through the sympathetic nervous system and the hypothalamic pituitary-adrenal axis to produce lipid abnormalities, high blood pressure, and clotting disturbances (Raphael, 2001). 1 1 Women's paid work has been treated as an additional rather than a primary role: this reveals an unstated, stereotyped assumption of a traditional unpaid role for women as homemakers. Few studies have attempted to treat work and home conditions symmetrically, although those that do have found that work conditions may be equally or more important for women's health than for men's (Macintyre and Hunt, 1997). It is not only women's experiences that have been poorly represented in socioeconomic classifications. Non-White women and men have fallen into the catch-all category "ethnicity." The characteristics generalized from the word are unclear and subject to many interpretations. It makes little sense to categorize as "Asian" people with distinctive histories, geographic origins, and cultures including Cambodian, Chinese, Filipino, Japanese, Korean, Laotian, Thai and Indian 53 Canadians. As well, people who recognize multiple ancestries may not identify with a single culture or geography, and an individual's identity may shift in response to the way the question is asked or with the purpose of the survey (Jones, Snider, and Warren, 1996). 1 2 The G N P is a misleading measure of wellbeing because it tells us little about the quality of our health and our lives. In Canada over the last 50 years the G N P has risen while many components of living capital (social cohesion, human health, environmental health) are worsening. Anielski (2001) proposed the Genuine Progress Indicators (GPI) to take a holistic, integrated, and living-capital-systems approach to measuring the conditions of our economic lives, social cohesion, human health, and environmental health. The GPI includes the full monetary cost benefit accounting of economic growth by considering the "value" of unpaid work (volunteerism, domestic work) and the costs of crime and environmental degradation (Anielski, 2001). 1 3 Low income refers to the Low Income Cut-offs identified by Statistics Canada. These cut-offs define low income in relative terms, based on the percentage of income that individuals and families spend on the basic needs of food, clothing, and shelter in comparison with other Canadians. Families and unattached individuals with "low incomes" usually spend more than 54.7% of their income on food, shelter, and clothing (Health Canada, 1999). The category identifies those who are substantially worse off than the average Canadian and are living in dire circumstances (Raphael, 2001). 1 4 According to Health Canada (1999), in Canada women aged 18 to 24 and aged 70 and older had the highest incidence of low income in 1990 and 1995. This report also found that overall men earned 62% more than women in 1995 ($31 117 versus $19 208). The largest wage disparity was found between men and women aged 55 to 64, where men earned almost double what their female cohorts did. In fact, men age 25 and older earned from 41% to 97% more than their female cohorts in 1995 (Health Canada, 1999). 1 5 Poverty is a normative concept that can be defined — in both absolute and relative terms — in relation to "need," "standard of living," "limited resources," "lack of basic security," "lack of entitlement," "multiple deprivation," "exclusion," "inequality," "class," "dependency," and "unacceptable hardship" (Krieger, 2001). 1 6 In a review paper on the research on health of low-income Canadian women, the following SES indicators were used: employment sector, social assistance recipient, income decile, family income, per capita income, perception of financial strain, education, life-circumstances index, low-income geographic area, subsidized housing, personal income, employment status, income quintile (geographic area), occupation, index of possessions, median income (geographic area) (McDonough and Walters, 2001). 1 7 For example, concepts of individual control over existing life circumstances might be a higher order variable that can synthesize a number of factors. Individuals higher on the SES ladder may have more frequent or more significant opportunities to influence the events that affect their lives, compared with people at the lower levels. This sense of control could affect education, occupation, housing, nutrition, health behaviours, medical care, and other aspects of social class experience not previously discussed (Adler et al., 1994). 1 8 Mortality (life expectancy) is the number of years one can expect to live. Infant mortality rates mean of 1000 infants born alive, how many die before living one year (Raphael, 2001). Morbidity rates measure incidence of disease. 1 9 Currendy there are no universal definitions of health. Perceptions vary across individuals and cultures depending on the meaning and importance people give to it. Reviewing the literature revealed three major themes regarding health - health as the absence of disease, a functional capacity, and a positive condition (Colantonio, 1988). Generally, conceptualizations of health range from narrow to broad — from the physical body to the ecosphere — and use terms such as biological, psychological, social, societal, environmental, cultural, economic, and political (Rootman and Raeburn, 1994). Jones and Meleis (1993) conceptualize health as empowerment. According to them health is a process of growth and becoming, of being whole and maximizing the development of one's potential. Health is being empowered to define, seek, and find conditions, resources, and processes to be an effective agent in meeting the significant needs perceived by individuals (Jones and Meleis, 1993). Wellbeing involves improved quality of life, efficient functioning, the capacity to perform at more productive and satisfying levels, and the opportunity to live out one's life span with vigor and stamina (McAuley and Rudolph, 1995). The "emerging consensus among researchers is that the term 'wellbeing' implies an emphasis on the individual's perception or sense of wholeness of self, groups, or community" (Schlicht, 1993). Health is an important component of wellbeing, but it is not the only component (Deaton, 2002). 54 2 0 A prevailing ideology during the early years of the women's health movement was that health concerns were so fundamental to women that they cut across race and class lines. Health was seen as a powerful link that could serve to unite all women into a strong and unified social movement. As a result of the criticisms leveled against white, middle-class feminists for generalizing the needs of dominant groups of women to all women, race and class were identified as the second and third "axes" of domination. Extensive theorizations about the "additive," "multiplicative," or "interwoven" nature of the gender, race, and class triumvirate resulted. More recently, however, some feminist researchers argue that "any naming or categorizing tends to call attention to similarities and to neglect differences, and any human or social phenomena can be understood in countless different ways" (Martin, 1994, p.643). Although feminists affirm diversity, it remains difficult to be certain that it means race, gender, and class to all women. How do we know that diversity does not mean being fat, religious difference, involvement in an abusive relationship, disadvantage at the workplace, or decisions made by girls in high school that attract them to female-dominated, lower paying jobs? The very categories we have assumed a priori (race and class) to be definitive of our differences may in fact be less significant than some others (Martin, 1994). For my purposes, I am focusing on urban Canadian women living in poverty. In this dissertation I am unable to fully consider the various other ways that the women may understand and experience their diversity. 2 1 I use the terms "unhealthy," "ill-health," and "the unhealthy" to represent the experiences of declining health, disability, and disease. Although I consciously use these terms as "catch-alls," in many cases the experiences of health, disability, and disease are distincdy different. It is beyond the realm of this dissertation to explore the particular ways that these experiences are different. 2 2 Stigmatization captures what Swanson (2001) calls "poor-bashing." Poor-bashing occurs when people who are poor are humiliated, discriminated against, shunned, pitied, patronized, ignored, blamed, and falsely accused of being lazy, drunk, stupid uneducated, having large families, and not looking for work (Swanson, 2001, p. 2). To stigmatize is to characterize or brand as disgraceful or ignominious. A stereotype is a conventional, formulaic, usually oversimplified conception, opinion, or belief (Morris, 1982). It denotes something fixed or lacking in originality. In common usage stereotyping can be seen as an ideological discursive strategy which demarcates an us/them binarity which functions to reinforce the domiant discourse (Gamble, 1999). To be invisible means to not be easily noticed or detected; to be inconspicuous (Morris, 1982). 2 3 "Culture is a matrix of beliefs, values, and norms that inform, give meaning to, and regulate experience" (Westkott, 1998, p.816). According to Young (1990), culture includes symbols, images, meanings, habitual comportments, stories, and so on through which people express their experience and communicate with others. The symbolic meanings that people attach to other kinds of people and to actions, gestures, or institutions can often significandy affect the social standing of persons and their opportunities (Young, 1990). 2 4 It is important to note that not all authorities are paternalistic and discriminating. The majority of welfare and health care workers, whether or not they agree, are forced to comply with their insitutional policies. Some attempt to resist and to use covert practices in order to help the poor. 2 5 For clarity and consistency I use the term participatory research (PR). Participatory research (PR) includes the sister trends of action research, participatory action research, and other schools of participative inquiry, including co-operative inquiry, participatory rural appraisal, and participatory evaluation (Greenwood and Levin, 1998; Maguire, 2001). 2 6 "Feminisms" is a term used to convey the diversity within feminist theoretical and political views on understanding the oppression of women (Ristock and Pennell, 1996). 2 7 Marx and neo-Marxists linked the concept of praxis with a production paradigm in the interests of historical explanation. In more recent times praxis has played a prominent role in the formation of the school of critical theory, in which praxis is seen to be more directly associated with discourse, communication, and social practices (Audi, 1995). 2 8 Integrative feminism, which recognizes diversity, is central to theoretical discussions on FAR (Maguire, 2001). Multiple inequalities exist on the basis of race, class and other factors (Cancian, 1992). Feminists have clarified how the concept of 'women in general' falsely universalizes and privileges the perspective of middle-class, heterosexual white women and denies and devalues the experiences of other women (hooks, 1984). Gender and women's diversity should be central in emerging explanatory frameworks. Hearing and affirming a multiplicity of voices can 55 'bridge the gaps' between people from different social positions (Ristock and Pennell, 1996). 2 9 On a personal note, at several academic conferences I have been warned that feminist action research is risky for graduate students, that I may encounter opposition in doing it, and that only tenured academics have the "liberty" or "freedom" to engage in such work. 56 Research Context and Methods Everybody gets together and basically is given the opportunity or the chance to express their concerns or views Or problems, it could be personal or whatever. And suggestions and ways and means of problem solving. Like what you can do about it. Or where you can go to get help (Joanne's interview, June 9, 2000). Over time social, pohtical, cultural, and economic factors are sohdified and reified into a series of structures that are taken as real, natural, and immutable (Guba and Lincoln, 1994). These factors are constructed with particular interests and goals in mind, and often they require critique and change. Both my ontological and epistemological understandings were linked to elements of critique and hope — my research framework centered on the experiences of a group of marginalized women, required dialogue between the researched and the researcher, and was action-oriented. Feminist action research (FAR) guided this dissertation because it fostered a critical understanding of women's multiple perspectives and promoted inclusion, participation, and collective action. Through being deliberately reflexive, as a feminist action researcher I attempted to remain as self-conscious and transparent as possible in the research process. While the ultimate goal is critiquing and demystifying power relationships and working towards social change, feminist action researchers do not try to produce universal or objective accounts; rather they believe that having partial and local knowledge that values women's experiences, autonomy, and diversity is still knowing (Tom, 1997; Richardson, 2000). Not only do feminist action researchers promote inclusion, participation, and action, but they emphasize researcher 57 reflexivity and transparency that can iUuminate and possibly untangle qualitative research's current "crises of representation and legitimation'' (Lincoln and Denzin, 2000). What follows is a description of this dissertation's context and methods and how, in it, my ideals for FAR were both upheld and forfeited. This chapter begins with an overview of the research context - the W O A W organization - and why, how, where, and with whom I carved out my dissertation project. I then explain the data gathering methods, including the Research Team meetings, one-on-one interviews, participant observations, and fieldnotes. A detailed description of the analysis methods and issues of representation, power, legitimation, and trustworthiness follows. Finally, I conclude with an exploration of qualitative writing as a method of inquiry (Richardson, 2000) and the promises of collaborative writing projects for producing truly authentic texts. In chapter 4 I provide a more detailed description of the research participants stories and experiences. The Research Context: In Collaboration with WOAW A Brief History of "Women Organizing Activities for Women" (WOAW) I conducted my dissertation in a larger feminist action research project named "Women Organizing Activities for Women" (WOAW). In June 1999 representatives from community recreation, social services, community schools, a women's centre, and a researcher from U B C (Dr. Wendy Frisby) collaboratively organized a 6 hour workshop for women on low income. The goal of this workshop was to uncover barriers to poor women's involvement in community recreation. Over 85 women, including single mothers, middle-aged and older women, and recent immigrant women, attended. At the end of the workshop the attendees decided to continue meeting to discuss their barriers and possibilities for increased involvement in community recreation because they saw involvement as contributing to their health. They named the impromptu organization "Women Organizing Activities for Women" (WOAW). At that time 58 they divided into four W O A W subgroups, with the intention of meeting regularly as subgroups and planning activities and workshops for themselves and women in their immediate communities.1 There were 10 different service providers2 affiliated with WOAW, and each subgroup had at least one service provider affiliated with it who worked as a resource and support person. The " W O A W Project Team" was a layer of the organization where all W O A W members — mcluding subgroup members, service providers, and UBC researchers — met. The Project Team fostered communication and networking among the four subgroups and developed important resources and relationships with community organizations. Generally the subgroups met every two weeks while the Project Team met once a month. This formation of the Project Team and the subgroups remains, though one of the original subgroups of primarily immigrant women disbanded in November 1999, while a new group, CoPoMo WOAW 3 , formed in the East Coqutilam/Port Moody area in April 2000.4 In September 1999 I became involved in W O A W when Wendy Frisby was completing a research proposal to the Social Sciences and Humanities Research Council (SSHRC). Though funding was not secured, Wendy was committed to supporting and researching WOAW, and with the remaining funds from a SSHRC seed grant5, she hired me as the Project Manager. I began attending W O A W subgroup and Project Team meetings in order to observe decision-making processes and to write fieldnotes. I also volunteered to write a monthly newsletter that was distributed to all W O A W members to foster communication and collaboration. During the fall of 1999 I met individually with most of the community representatives and with some of the women on low income to answer their questions about the research grant, the possibilities for the funding, and other general inquiries. At that time my role was to maintain a UBC presence in W O A W while budding trust and making research, the grant application process, and the role of 59 university-based researchers as transparent as possible. In December 1999 Wendy received 3 years of funding from SSHRC to study WOAW. 6 The overall goal of the research proposal was to examine the health problems of physical inactivity, stress, and social isolation through a feminist action research process designed to increase the involvement of women on low income in community recreation. The specific research objectives of the project were: i) to examine the lay meanings that physical inactivity, stress, and social isolation have for this diverse gtoup of women on low income; ii) to examine the models of feminist action research that emerge as the 4 groups develop action plans to address these self-determined health problems; and iii) to develop an initial framework that would inform the evaluation of community recreation interventions for women on low income. Refer to Appendix A for an overview of the 3 year SSHRC project. The SSHRC research proposal referred to W O A W as a feminist action research project within a collective organization. Although some W O A W members had read the SSHRC research proposal and knew we identified our research as feminist action research, they rarely referred to FAR direcdy. As an organization W O A W had the primary goal of reaching women with financial, social, cultural, or physical barriers to participate in their communities. W O A W members, with the guidance of a two service providers, decided to organize as collaboratively and inclusively as possible. The W O A W members deterrrrined that in order to redress poor women's isolation and alienation from many community services, W O A W should function as a collective organization. According to the two service providers, collective organizations made decisions through consensus, had a flat organizational structure, and encouraged equal participation at all meetings. There were ongoing challenges, however, with group process at all levels of organizing, and many W O A W members were confused and frustrated by consensus decision-making and did not clearly understand the differences between a collective organization 60 and a more traditional hierarchical one. Most W O A W members had httle or no experience working in a collective and had never made decisions through consensus. Misunderstanding the collective process was evident when Katharine, a woman on low income, showed me a diagram that represented her understanding of WOAW. Beth, a UBC graduate student, wrote in her fieldnotes that the diagram Katherine drew of W O A W for the newsletter "is hierarchical, with Wendy at the top, followed by Colleen, the Project Team, and then the subgroups. Sydney [a UBC graduate student] and I are at the bottom; she called us Wendy's secretaries'" (Beth's fieldnotes May 15, 2000). Although this was not the only time that there were clear differences between the ideals of W O A W and how some W O A W women understood and experienced it, overall there was support for organizing as a collective and determining ways to facilitate everyone's participation. In the year 2000 the Project Team met several times to discuss WOAW's vision and values and the challenges of organizing as a collective. These meetings fostered a commitment to developing a greater understanding of collective organizing. As well, W O A W developed a vision statement: Women Organizing Activities for Women (WOAW) is diverse women working together to enhance quality of life and create positive and sustainable change. Women are empowered, respected, and connected to their communities. Al l thoughts and feelings are valued and important, and women are treated with dignity (WOAW vision statement, written collectively on May 10, 2000). Although W O A W members identified W O A W as a collective organization, at times "community development" was also used to describe the work of WOAW. Several service providers adopted community development discourses because it was consistent with their organization's rhetoric for increasing capacity in the local communities. In fact, community development was an espoused mandate for aU of the service providers who worked in the local parks and recreation departments. Community development can be understood as a process of 61 engaging individuals and organizations in collective action towards social change (Seebaran, 2000; cited in Millar, 2002)7; its ideals are consistent with both collective organizing and feminist action research. "Collective organization" and "community development" were used interchangeably by many of the W O A W members, and differentiating them was never raised as a need or concern. For the purposes of my dissertation I refer to W O A W as a collective organization.8 Important WOAW Phraseology Since my dissertation occurred within W O A W it is important to delineate the people involved and the organizational terms that were used. W O A W Members: The People Involved in W O A W Women on low income: WOAW women refers to a diverse group of women in the Tri-City area who had limited incomes, were socially isolated, and encountered barriers in accessing community services. During the three year SSHRC research project there were between 40 and 80 women on low income involved with WOAW. Throughout this dissertation research participant(s) refers only to the W O A W women who were actively involved in the dissertation research project as part of WOAW. The research participants were Susan, Elizabeth, Kelly, Maey, Caroline, Helen, Wanda, Rene, Martha, Arlene, Cynthia, Teresa, Cassie, Willow, Katharine, Alexa, Joanne, Trina, Vhginia Dawn, Gloria, Cliristine, and Julie. Most of the women chose to use pseudonyms; this is discussed later in this chapter. Service Providers: refers to people who had paid employment in the Tri-City Area (Coquidam, Port Coquidam, Port Moody), were affiliated with WOAW, and through their paid employment supported W O A W's organizing. Between 10 and 14 recreation centre, women's centre, community school, and social service workers were service providers. I called them service providers since they brought resources and support from their workplaces to support 62 WOAW. The service providers were active to varying degrees — some attended all subgroup and Project Team meetings, some attended some of the meetings, and others acted primarily as liaisons for the subgroups, facilitating the organization of activities, chudminding, room bookings, and general networking. The service providers were not research participants in this dissertation study and did not participate in the Research Team. Researchers: The UBC Working Group includes Dr. Wendy Frisby, Pam Ponic, Sydney Millar, Jo-Ann Zyla, Beth Pinnington, and myself. Wendy was on faculty in the School of Human Kinetics at U B C and we were graduate students who worked with her. Periodically other research assistants were hired to do specific work tasks and attend Working Group Meetings. From September 1999 to April 2001 I was the Project Manager and attended most W O A W meetings, wrote fieldnotes, coordinated data collection and analysis, managed the research budget, and established organizational systems for the research study.9 In May 2001 Pam Ponic became the Project Manager. The UBC Working Group met every 2 to 4 weeks at UBC to discuss WOAW, the SSHRC grant, and our roles as feminist action researchers. Most of us were involved in data collection for the SSHRC project, and Pam, Sydney, Beth, and I also gathered data for our own graduate work.10 Beth, Pam, and Sydney wrote fieldnotes that contribute to the data set of this dissertation. WOAW members refers to everyone involved in W O A W including the women on low income, the service providers, and the researchers. Although the above descriptors are useful for extinguishing the different group of people involved and their different reasons for involvement, in naming them it is possible to falsely categorize or falsely differentiate those involved. For instance, before their involvement in WOAW, two W O A W women were social workers, a researcher was previously a woman on low income, and one service provider had engaged in graduate work. These categories roughly describe W O A W members at the time of 63 their involvement in WOAW, and are not to be understood as fixed or static. Layers of the W O A W Organization WOAW (Women Organizing Activities for Women) refers to the collective organization that was initiated in June 1999 and includes both the organizing and the research components of WOAW. SSHRC project refers to research project that was funded by SSHRC (Social Sciences and Humanities Council) to study the W O A W organization and women on low incomes' primary health concerns. The SSHRC research budget supported all research-related activities, some W O A W activities, and research honoraria.11 In order to manage the SSHRC research budget, Wendy divided the three-year study into six phases with each phase six months in duration (i.e. Phase 1 — January to June 2000; Phase 2 — July to December 2000; and so on). She allotted money for each phase and asked that one Project Team meeting per phase be devoted to discussing the budget. Although the phase budgets and honoraria varied over the course of the SSHRC project, the collective principles for determining the budget remained consistent. The WOAW Project Team met every 2-4 weeks; any person who was involved in W O A W was invited to attend the Project Team meetings. On several occasions, people who were interested in W O A W and in possibly starling a W O A W group in their local community attended Project Team meetings in order to learn more about WOAW. Typically there were between 15 and 25 people in attendance at the meetings. A variety of issues were discussed at these meetings, mcluding subgroup activities and updates, upcoming events, resource sharing, relevant community issues, challenges and conflict with group process, and the SSHRC research project. At each meeting a facilitator and a minute-keeper volunteered; these positions rotated at most meetings. The majority of the W O A W women and service providers were affiliated with a WOAW 64 subgroup. Some W O A W women were involved exclusively in their subgroup while rarely attending other levels of organizing in W O A W , while other W O A W women were more active in the Project Team and Research Team. Although membership and activities varied over the course of the SSHRC project, the subgroups were as follows: Aged to Perfection (ATP) formed in June 1999, and 20 to 25 women and 1 service provider were active in this group. The majority of A T P members identified themselves as seniors, though the group did not impose strict age requirements. They worked primarily out of Dogwood Pavilion and Hoy Creek Housing Co-Operative in Coquidam and organized activities such as belly dancing classes, tai chi, computer framing, and "Safety Day". CoPoMo WOAW formed in Apr i l 2000, 15 to 20 women, primarily middle-aged and recent irrimigrant women, and 4 service providers were involved. The meetings and activities occurred at three locations in Coquidam - the Coquidam Aquatic Centre, Pinetree Community Centre, and Hoy Creek Housing Co-Operative. They organized activities such as career planning, fitness training, and hosted a series of workshops mcluding one on "Stress Management." PoCo WOAW formed in June 1999 and 6 to 12 women and 2 service providers were affiliated with this group. Meetings and activities began in the Port Coquidam Area Women's Centre Society, though they moved to Hyde Creek Recreation Centre in Apr i l 2000. The majority of PoCo W O A W women were young mothers with dependent children. This group organized activities such as qi gong, tai chi, yoga, mom and tots, weight trairring, aerobics, First A i d , and computer classes. SWCo WOAW formed in June 1999 and 5 women and 3 service providers met regularly. They organized activities such as a walking group, health cooking classes, and an English discussion group, and held most of their activities and meetings in Mountainview Pubhc School and Poirier Recreation Centre, Coquidam. They were also active in their community, making 65 connections with groups such as a single mothers' support group and a multi-cultural society. This group was diverse - they were a variety of ages and some had young children, grown children, or no children. Table 1 - The Women's Subgroup Affiliations WOAW subgroup affiliation Subgroup size Names of research participants Aged to Perfection 20 to 25 Rene, Wanda, Arlene, joy, Trina CoPoMo W O A W 15 to 20 Katharine, Alexa, Cassie, Teresa PoCo W O A W 6 to 12 Kelly, Virginia Dawn, Julie, Joanne, Caroline, Cynthia SWCo W O A W 5 Susan , Willow, Maey, Elizabeth, Gloria Note: Although some women moved between subgroups, when I wrote my dissertation the 20 women involved as research participants belonged to these subgroups. The Research Team first met in March 2000. I initiated the Research Team as a place for the W O A W women to discuss common issues and concerns, particularly those associated with their health and living in poverty. Throughout my dissertation I refer to the women who regularly attended the Research Team meetings as Research Team members.12 I attended most of these meetings and often another researcher attended to record fieldnotes. Only once, when invited by the Research Team members, did a service provider attend. Between 1 and 17 women attended, with an average attendance of 11 women. Although these meetings were the primary data source for my dissertation (from March 2000 to June 2001) the meetings and my involvement in them continued and is ongoing today. My involvement is further explained later in this chapter. Research Parties occurred every 6 months (corresponding to phases of the SSHRC project) and were organized for the UBC researchers to meet with the W O A W women to review research findings or to discuss ongoing issues in WOAW. At these 3 hour meetings the SSHRC grant provided a catered lunch and an honorarium for women in attendance. At the Service Providers' Research Parties the researchers met with the service providers and discussed issues and concerns that they were encountering. Most of these meetings had an open agenda and the service providers raised pertinent issues — for instance, at one meeting we discussed the 66 power issues and dynamics that were surfacing at different levels of WOAW. Consult figure 2 for a graphic representation of the W O A W organization and its members. Figure 2 - The WOAW Organization 67 Conducting F A R Collaboratively There was a seamless fit between W O A W as a new grassroots collective organization and my research interests. From the beginning of my involvement as the Project Manager, I intended to carve out my dissertation research from within the W O A W organization. In November 1999 I met with Louise, a W O A W service provider, who articulated a need for a "research perspective" in her subgroup because discussions of the "politics of poverty" had not yet occurred. At that time there was no place in W O A W for the women to discuss major life issues and constraints as the focus of Project Team meetings was on activity planning and resource sharing. According to her, through discussing the politics of poverty the women could share experiences, better understand their poverty and health, and possibly engage in collective action. Together we brainstormed and determined that the creation of a "Research Team" could fill this void and be an ideal site for collecting data for my dissertation project. The first Research Team meeting occurred in March 2000. In Maguire's (1993) reflections on participatory action research, she states that "it may be easiest and most instructive to try becoming involved in an ongoing or established participatory research project in which you can contribute your work without having to mobilize the entire project from scratch" (Maguire, 1993, p.175). Collaborating with the SSHRC project and working as the Project Manager facilitated all phases of my dissertation project. The challenges of access, start-up, tmst-bunding, resources, and support were better handled through the larger FAR project since there was a budget for research activities and a number of researchers, research participants, and service providers involved. Nevertheless, the number of researchers involved and the large SSHRC budget, collaboration and power dynamics were complex and multi-faceted. As both a Ph.D. candidate and the Project Manager I worked with Wendy Frisby as one of my research committee members and also as my "boss." My research advisor, Allison 68 Tom, was a collaborator on the SSHRC project and in the second year initiated her own research project to study the challenges of feminist action research. Given that one of FAR's goals is to work collaboratively, these power dynamics reflected the reality of conducting FAR and provided insights into the challenges and potential contributions of working collaboratively as feminist action researchers with community groups. Despite these complex power relationships, conducting my research within the W O A W organization and the SSHRC project had many benefits. The relationship between the Research Team, WOAW, and the SSHRC research project enabled the Research Team meetings to evolve as they did. From the begmning the Research Team explored personal and emotional issues that had not been raised at other meetings witiiin WOAW. Susan explained: My idea of the Research Team is very different than the other groups in the sense that it has its own agenda and to me it's been the fact that we're taking particular issues that are vitally important to the individual, as well as the groups, 'cause I noticed there's a lot more personal conversation going on here than in a lot of the other groups (Susan, R T M October 19, 2000). At the Research Team meetings we focused on the women's more personal issues because the subgroups and Project Team handled WOAW's "business", including membership, the research budget, and organizing activities and workshops. As well, to some degree the structure of W O A W allowed the women to choose how and where they wanted to be involved. Maey said "And one group doesn't work out, just go 'ok, this isn't the group for me, I'm going to try the other group and see how this group works for me cause every group is different'. Every group is a lot different" (Maey, R T M October 19, 2000). The layers of the W O A W organization enabled some flexibility and choice for the W O A W participants and possibly helped them to maintain their energy and enthusiasm for the Research Team meetings.13 Not only did the W O A W organization provide different avenues for the women's involvement, but, significandy, the SSHRC grant covered important costs that facilitated the 69 women's involvement in all aspects of WOAW, mcluding the Research Team. As well, the SSHRC grant provided me with resources to facilitate my own research, such as covering my own travel expenses and paying for a researcher to record fieldnotes, transcribe interviews, and watch the women's young children during one-on-one interviews. As a doctoral candidate and neophyte feminist action researcher, I was in the enviable position of having the support of W O A W and the SSHRC grant to pursue my research interests. Data Gathering Given the methodological limitations and dearth of qualitative studies in the health equity field as described in chapter 2, qualitative methods were appropriate for this study. The use of qualitative methods locates the individual in the context of social structures and can broaden the range of debate. In the qualitative arena the individual is not only inserted into the study, the individual is the backbone of the study. The description of persons, places, and events has been the cornerstone of qualitative research (Janesick, 2000). "Qualitative research is conducted not to confirm or disconfirm earlier findings, but rather to contribute to a process of continuous revision and enrichment of understanding of the experience and form of action under study" (Elliott, Fischer & Rennie, 1994; cited Lincoln, 1995). Qualitative methods were chosen because of their potential to show how broader structures constrained and shaped poor women's lives, while enabling them to describe these influences. Poor women are often excluded from knowledge production processes and qualitative methods legitimize their experiences and let their voices be heard. Research Questions I posed questions in three areas: • How did a gtoup of women on low income understand and negotiate their experiences of poverty, exclusion, and health, both individually and collectively? 70 • From the perspectives of a group of women on low income, what was the connection between poverty and health? In what ways did they believe that the experiences of exclusion, stereotyping, and ^visibility affected their health? • How did a group of women on low income collaboratively identify, organize, and enact action towards social change to address their shared concerns? The findings from these research questions are discussed in chapters 6, 5, and 7 respectively. Only a fraction of the qualitative data that were gathered for the SSHRC project were used to examine my research questions.14 These data are represented in Table 2. Table 2 - Overview of SSHRC and Dissertation Data Subgroup meetings Project Team meetings Research Team meetings Interviews W O A W women Interviews Service providers Working Group meetings Research Parties Phase 0 Sept-Dec 1999 Fieldnotes Fieldnotes Fieldnotes Phase 1 Jan-June 2000 Fieldnotes • Fieldnotes Transcripts Eieldnotesfc:; 1 ranscripts . Transcripts 3 -'schedule 1 - schedule 2 Transcripts Fieldnotes Transcripts '.Fieldnotes Transcripts Phase 2 July-Dec 2000 Fieldnotes Fieldnotes Transcripts Fieldnotes; Transcripts Fieldnotes Transcripts Eieldridte! Transcripts Phase 3 )an-June 2001 Fieldnotes Transcripts Fieldnotes Transcripts • Fiel3h6tes|:»-Transcripts Transcripts Transcripts Fieldnotes Fieldnotes Transcripts Phase 4 July-Dec 2001 Fieldnotes Fieldnotes Transcripts Fieldnotes Transcripts Fieldnotes Transcripts Phase 5 Jan-June 2002 Fieldnotes Transcripts Fieldnotes Transcripts Fieldnotes Transcripts Phase 6 July-Dec 2002 Fieldnotes Transcripts Fieldnotes Transcripts Fieldnotes Transcripts #Note:; Datafused f^my.'disserStiom| a There were two "rounds" of interviews in this phase of the SSHRC project. Refer to Appendices E and F for the interview schedules. As can be seen in Table 2, the major data sources for my dissertation were transcripts from the Research Team meetings, the Research Parties, and two rounds of interviews, and fieldnotes. As I explain later in this chapter, I remained connected to the Research Team throughout the SSHRC project though I did not use all of its data for my dissertation. Throughout my dissertation the data are identified according to Table 3. 71 Tab le 3 - Key for R e a d i n g Quo ta t i on s Type of Data Example Interview data Willow's interview, September 14, 2000 Research Team meeting (RTM) data Willow, R T M June 29, 2000 Fieldnote data Colleen's fieldnotes, October 19, 2000 The Women's Involvement as Research Participants Al l of the women who participated as research participants in my dissertation were W O A W members. Involvement in my dissertation was entirely optional — some W O A W women were consistendy involved throughout my project, some were involved in the begmning and ceased their involvement, others became increasingly involved towards the end, while other W O A W members were never involved. The majority of the data used throughout this dissertation are based on the 20 women who were initially involved in the Research Team and who participated in the first round of interviews. Over time, 5 of these women withdrew from the Research Team, while another 10 women became involved to varying degrees. Only one woman who was interviewed in the first round never attended a Research Team meeting (she was active in other aspects of WOAW). A n overview of the women's involvement in my dissertation appears in Table 4. The women who were involved in my dissertation were diverse in terms of their ages, educational backgrounds, and work experiences. They all identified themselves as "poor," and the majority of them received social assistance, cUsability benefits, or a seniors' pension. Two-thirds of the women were mothers and identified as single. In chapter 4 I provide detailed accounts of the women's backgrounds and experiences. 72 Table 4 - Overview: WOAW Women's Involvement in Dissertation • Average number of women in attendance at the Research Team meetings 11 • Number of different women who attended the Research Team meetings Number of women who attended 1-3 RTMs Number of women who attended 4-6 RTMs Number of women who attended > 7RTMs 30 13 4 13 • Number of different women who were interviewed Number of women interviewed in 1 s t round Number of women interviewed in 2 n d round 20 20 12 • Number of women who were interviewed but never attended a Research Team meeting 1 • Number of women who were interviewed but who, as of December 2000, were no longer involved in W O A W or the Research Team 5 • Number of women who became involved in the Research Team after July 2000, were not interviewed for the purposes of this dissertation15, but who signed consent forms and contributed to this dissertation's findings by participating in Research Team meetings 11 Chronology and Overview of Data Collection This section provides a general overview and chronology of the data collection. A detailed description of the data collection activities, deluding the Research Team meetings, the one-on-one interviews, and the participant observations, follows this section, and in chapter 7,1 analyze the Research Team as feminist action research. For more detailed information, consult Appendix A, "Overview of the SSHRC Project" and Appendix B, "Detailed Overview of Research Team Meetings." Gaining Access and Building Trust: September 1999 to February 2000 From September 1999 to March 2001, I worked as the Project Manager, identified the Research Team as the primary site of my data collection, received committee approval for my "second stage" research proposal, received UBC ethical approval, and generated interest in the Research Team meetings. Maguire referred to this stage as "entering, experiencing, estabUshing relationships with actors" (Maguire, 1987, p.116). During this time I attended all W O A W meetings as a participant observer and recorded fieldnotes. I became acquainted with most of the W O A W women and service providers and learned about the Tri-City area. 73 Intensive Data Collection: March 2000 to June 2001 March and April 2000: Orientation to the Research Team There were three Research Team meetings between March and April 2000. At these meetings we discussed the purpose of the Research Team, the role of research, general expectations and guidelines for meeting, and issues around informed consent, tape recording, anonymity, and confidentiality. The women were not concerned about tape recording the meetings and at times they reflected that tape recording was important because it facilitated the research. During this time the Research Team resembled group interviews more than group meetings. I asked the women questions about their experiences kving in poverty, their major health issues, and the relationship between poverty and health. I pursued these topics in a first round of one-on-one interviews that occurred in March and April 2000. The first three Research Team meetings were attended by a variety of different women; some of whom became regular attendees while others did not. In these early meetings I was uncertain whether the women would choose to continue to meet as the Research Team. May and June 2000: Anticipation and Celebration Energy and enthusiasm for the Research Team grew in the spring and summer of 2000. New members continued to join the Research Team while a core of Research Team members developed. By May 2000 the Research Team more closely resembled a meeting place with a core group of members than a group interview. The Research Team members seemed increasingly familiar and comfortable with the purpose and structure of the meetings, and began to ask their own questions. During this time the women shared a sense of celebration for what they had accomplished as the Research Team and the W O A W organization. In June I conducted the second round of interviews with 12 women and asked them about 74 the Research Team. Katharine said: [The Research Team is] very big. A big aspect of WOAW. We don't want it to stop. I want it to grow. I want it to have more power. We need to come to a point where we can make recommendations or at least explain some facts. And then approach the community and say "look at things that are not done" (Katharine's interview, June 9, 2000). Generally the women shared a sense of collective power and believed that the Research Team played an important role in WOAW. Wanda commented "it's [the Research Team] the most important thing. Without the Research Team I don't think the rest would gel together.... I don't see W O A W itself as being held together without that Research Team" (Wanda's interview June 5, 2000). The women felt that the Research Team engaged in relevant discussions, and that the actions they had identified were important to pursue. August to December 2000: Work and Conflict After the high energy of the spring and early summer of 2000, the Research Team members chose to not meet in July and to reconvene in August. The meetings that followed were riddled with conflict. There was litde evidence of the connection and closeness that characterized the previous stage of the Research Team. The members experienced an increasing need to do work and to be productive, and a decreased value in exploring issues and sharing as a gtoup. In September 2000 the women identified that they wanted to work towards registering W O A W as a non-profit society and coordinating a community forum to address their grievances with the welfare system. With the Research Team's impetus for doing work, meetings were lengthened to 3 hours with half of every meeting devoted to each action item. Once this decision was made conflict arose. The Research Team members were increasingly frustrated and at the November 2000 meeting only one woman attended. At this point I too was frustrated — I felt that there was little energy for the Research Team and decided to not initiate another meeting. During this time I recorded my emotions and reactions in my fieldnotes and journal. 75 The UBC Working Group organized the December 2000 Research Party in order to collect data on the women's experiences in the first year of the SSHRC project (2000). Although 12 women attended they were clustered into small groups around the table. There was antagonistic "side-talking" throughout the meeting and tension between the attendees. January to June 2001: Renewal of the Research Team In mid-January 2001 I received an email from Wanda inquiring about the Research Team and our next meeting. She asked to schedule a Research Team meeting and suggested that six women she had consulted would attend. I then organized a Research Team meeting for February 6, 2001, and 9 women attended. We discussed our previous disappointments and strategies for managing our shared frustrations with the Research Team meetings. The woman determined that the Research Team would meet once a month, for 2 hours, and would only focus on the welfare grievances workshop. We then had monthly meetings, had a W O A W service provider attend, and hosted a workshop with D E R A (Downtown Eastside Residents' Association) to gather information on B.C. Benefits. Continued Researcher Involvement and Commitment: July 2001 to the Present Although I completed my data collection in June 2001, I remained committed to and involved in W O A W and the Research Team. I was no longer the Project Manager and therefore met with W O A W members less often. However, I continued to coordinate and attend Research Team meetings, attend UBC Working Group meetings and the occasional Project Team meeting. During this time I corresponded with some W O A W members in order to devise strategies for communicating the fmdings from my dissertation to the W O A W organization and the broader community. 76 Data Sources The Research Team The Research Team meetings' transcripts and fieldnotes were the largest data source for my dissertation. There were many benefits in estabhshing a group situation as my primary data collection method. According to Madriz (2000), groups are particularly suited for feminist research because they minimize the control of the researcher during the data gathering process by decreasing power over the research participants and limiting the control of the moderator, who has less power over a group than over a single individual. As well, groups emphasize the collective, rather than the individual, and can foster free expression of ideas while encouraging the members of the group to both recall experiences and to speak. Group meetings can allow moments of not having to talk, of being able to listen to others, thus enabling rethmking initial accounts that, upon reflection, may need amplification, qualification, amendment, or contradiction (Lofland and Lofland, 1984). Researchers working with groups can witness participants engaging in dialogue, sharing ideas, opinions, experiences, and even debating with each other. It is also possible to see the participants' own power relations and hierarchies (Madriz, 2000). However, group meetings can be a challenge for the researcher. At times it may be difficult to keep one person from dominating the group, to encourage reluctant respondents to participate, or to obtain responses from the entire group to ensure the fullest coverage on the topic. As the researcher and facilitator of group meetings it is often difficult to know what to do. In group meetings the researcher must balance the directive, interviewer role with the role of the moderator, which calls for the management of the dynamics of the group. The researcher must simultaneously worry about the script of questions and be sensitive to the evolving patterns of group interaction (Fontana and Frey, 2000). 77 Despite its challenges, group meetings were the most suitable primary data collection method for this feminist action research project since we were exploring experiences of poverty and health and working collaboratively to uncover action strategies. According to Randall (1980), groups can provide ways for women to face social isolation and oppression. Shared testimonies, individual and collective, can become a vehicle for capturing the socioeconomic, pohtical, and human challenges that women face (Randall, 1980; cited in Madriz, 2000). Groups are ideal for social researchers interested in building new paradigms of social research promoting social change (Madriz, 2000). The Research Team meetings were two or three hours long, occurred in Port Coquidam at Hyde Creek Recreation Centre, and all but two were tape recorded. I handled logistical issues such as booking the meeting room and the clindmmding, bringing snacks, and ensuring that all women who wanted to attend had transportation. I always picked up at least one woman for the meetings, and I often had a full car. These meetings, which occurred every three to five weeks, involved a considerable amount of my time. Every meeting required approximately fifteen hours of work, including meeting preparation and organization, transcribing the tapes of the meetings, writing fieldnotes, and travel. I characterized my roles on the Research Team as "researcher," "facilitator," and "Research Team member." These roles are examined in detail in chapter 7. The general activities of the Research Team included the identification and description of key health needs, concerns, and experiences; the exploration of common and divergent issues with poverty and B.C. Benefits; the identification of action plans and strategies; and the "work" of action. The Research Team evolved through the typical developmental stages of group formation, identification, and action. However, these stages were not linear nor were they discrete. In fact, action strategies emerged from the beginning of the Research Team. Many of the women who were involved were interested in engaging in the research component by 78 questioning and analyzing their experiences, observing and critiquing group processes, and working towards social change. Initially the Research Team meetings were semi-structured; at the first four meetings the women relied on me to ask questions to initiate discussion. As the Research Team gelled as a group and identified actions, the meetings became increasingly unstructured, though this was not always seen positively. The perceived need for structure in the Research Team meetings is further explored in chapter 7. Consult Appendix B for an overview of all of the Research Team meetings in terms of dates, attendance, and discussion topics. A n overview of the Research Team meetings is provided in tables 5 and 6. Table 5 - Research Team Meetings (March 2000 - June 2001) Research Team Meeting Data Collected DATE Transcripts Fieldnotes March 7, 2000 None - Consent forms signed a Colleen, Beth April 4, 2000 Primary transcript Colleen April 26, 2000 Primary transcript Colleen, Beth May 16, 2000 Primary transcript Beth June 7, 2000: 'Research Party Primary transcript Beth June 29, 2000 Primary transcript Colleen August 17, 2000 Primary transcript Colleen September 6, 2000 Primary transcript Colleen September 27, 2000 Primary transcript Colleen October 19, 2000 Primary transcript Colleen November 9, 2000 Primary (interview) transcript Colleen December 12, 2000: Research Party None - Tape recording blocked b Beth, Sydney February 6, 2001 Secondary transcript Colleen, Pam March 6, 2001 Secondary transcript Colleen, Pam April 3, 2001 Secondary transcript Colleen, Pam April 17, 2001 Secondary transcript Colleen, Pam May 29, 2001: Data analysispotluck Secondary transcript Colleen June 19, 2001: DERA Workshop Secondary transcript Pam Note: Primary transcript refers to tape recorded Research Team meetings that I transcribed and analyzed in my primary data set. Secondary transcript refers to Research Team meetings that were tape recorded and transcribed though the transcripts were not considered for my primary data set because of the volume of data already collected for my dissertation. By the end of the year 2000 I had a large data set, had reached saturation in addressing my research questions, and, with the encouragement of my research committee, decided to end the data collection for my dissertation. These secondary transcripts may be used at a later date by myself or the other research collaborators from the SSHRC project. a This Research Team meeting was not tape recorded since at this meeting the women signed consent forms and were familiarized with the process of tape recording, transcribing, and analysis. b Tape recording was blocked anonymously at this meeting. 79 Table 6 - Data from the Research Team Meetings • Total number of Research Team meetings 18 • Number of primary R T M transcripts 10 • Number of secondary R T M transcripts (not analyzed for this dissertation) 6 • Number of R T M that were not tape recorded (primary RTMs) 2 One-On-One Interviews Intensive interviews seek to discover the informant's experience of a particular topic or situation (Lofland and Lofland, 1984). One-on-one interviews are active interactions between two people leading to negotiated, contextually based results. Each interview context is one of interaction and relation; the result is as much a product of this social dynamic as it is a product of accurate accounts and replies. The nature of the social dynamic of the interview can shape the nature of the knowledge generated. Interview participants are "actively" constructing knowledge around questions and responses (Fontana and Frey, 2000). I conducted 32 one-on-one, semi-structured interviews, and they ranged in length from 45 minutes to 2 hours. Twenty women were interviewed in the first round; in these interviews I asked the women about their experiences with WOAW, their major health concerns, and specifically about stress, depression, social isolation, and physical inactivity. These interviews were used for both the SSHRC project and my dissertation, and in general I used two-thirds of the data from them to address my research questions. Before each interview I explained the nature of the research, the differences between my research project and the larger SSHRC project, and issues around confidentiality and informed consent. Together we read through the consent form (see Appendix C) and I explained why the UBC Ethics Committee requires them to sign such a form. I also explained the entirely voluntary nature of both my project and the SSHRC project. In the second round of interviews, 12 women identified as wanting to be interviewed; these 80 women had remained consistendy involved in die Research Team meetings and WOAW. In this round of interviews I asked the women about their health, the relationship between health and living in poverty, the experience of being stereotyped and stigmatized, and their reflections on the Research Team. Al l of these topics had been discussed at the Research Team meetings; I used the interviews to uncover individual experiences that may not have emerged in the group setting of the Research Team meetings. In the second round of interviews the participants' comfort and understanding of the research project and the interviews as a form of data collection had grown. By this time they had a clear sense that there were two distinct projects occurring and that the transcripts from the Research Team meetings and interviews were to be used as "data." Consult Appendices D and E for the interview schedules. I approached the interviews as an intensive guided conversation with the goal of eUciting rich, detailed, and personal reflections. Another goal of the interviews was to build rapport and reciprocity between myself and the women. I provided feedback to the women in a maximally reciprocal way, and did not presume neutrality. According to Oakley (1993), in feminist interviewing the interviewer refuses to stay detached and instead carries an obligation to reveal some of her own feelings in order to introduce greater reciprocity into the interaction. Women who lack opportunities to engage collaboratively with other women and women who are socially isolated may welcome the opportunity to talk to a sympathetic listener. But it is possible that the very effectiveness of these techniques leaves women open to exploitation through the research process (Finch, 1993). As much as possible I developed reciprocal relationships in terms of sharing my own experiences and personal information, though there were power and income differences between myself and the women I interviewed. I attempted to not minimize interviewer-interviewee differences and was honest and upfront about my interests in this research project and my personal background and differences. 81 Although the interviews provided an opportunity for the women to reflect individually on the discussions and group process of the Research Team, the research participants had different reactions to them. Oakely (1993) claims that the convention of the interviewer-interviewee hierarchy is a rationalization of inequahty and that what is good for the interviewer is not necessarily good for the interviewee. Indeed, in some interviews I realized that interviewing is an academic exercise that may raise insecurities and discomfort in some participants. After one interview, I wrote: She [research participant] said that she was uncomfortable [in the interview] because she didn't feel like she was answering the questions correcdy. She felt like she used to feel in school, when she was really worried about getting the right answers. She had purposely not asked me for the interview questions in advance because she wanted to practice answering questions without knowing what the questions would be. But she was nervous about the interview and struggled to find the "right" answers (Colleen's fieldnotes, June 2, 2000). Another participant said she had difficulty articulating her thoughts in an interview. "I have all these thoughts going through [my head] and I can't get them all to form... it's half there... half floating" (Elizabeth's interview June 12, 2000). Additionally, there were many instances where the participants ended their remarks with "I don't know," "you know?" or "I'm not sure if I'm answering the question." I was concerned about the comfort of the research participants when they suggested that they felt awkward or uncertain through the interviews. Some participants thought that there were right and wrong answers. These experiences conformed with Oakley's critique of interviewing, and I questioned what some of the women had gained from being interviewed. I had the opposite reaction from other research participants, however. At her first interview, Kelly greeted me at the door: "Colleen, I've been looking forward to this, I've got a lot of tilings I want to tell you" (Colleen's fieldnotes, March 14, 2000). She had made coffee and muffins and our interview was 2 hours long. Katharine began our interview with "So how does this work, do 82 I start or do you?" (Colleen's fieldnotes, March 14, 2000). She had made notes and had a list of things she wanted to raise with me. Through the interviews I was able to check for individual experiences, assumptions, and reflections. Importandy, I was able to talk in-depth with those research participants who were typically more quiet and withdrawn in group settings. Alexa said "I'm not an outspoken person. I'm not willing to share, I don't feel comfortable enough because I'm not used to being in the group" (Alexa's interview, June 2, 2000). The women who appeared the most uncertain about the responses they gave in the interviews were also the women who had difficulties expressing themselves at the Research Team meetings. I cannot assess whether the discomfort they expressed stemmed from their involvement in the research alone or whether it is something they encountered in other contexts. Tables 7 and 8 provide an overview of the interview data. Table 7 - One-on-one Interviews (March - September 2000) Interviewee Date(s) Alexa March 2, 2000; June 2, 2000 Arlene April 4, 2000 Caroline April 12, 2000 Cassie April 18, 2000 Cynthia April 12, 2000 Elizabeth March 20, 2000; June 12, 2000 Gloria March 24, 2000 Joanne March 20, 2000; June 9, 2000 Julie March 14, 2000 Katharine March 14, 2000; June 9, 2000 Kelly March 14, 2000; June 6, 2000 Maey March 27, 2000; June 9, 2000 Martha April 4, 2000 Rene Match 15, 2000; June 9, 2000 Susan March 24, 2000; June 2, 2000; November 9, 2000 Teresa March 20, 2000 Trina March 21, 2000; June 5, 2000 Virginia Dawn March 14, 2000; June 6, 2000 Wanda March 15, 2000; June 5, 2000 Willow March 15, 2000; September 20, 2000 83 Table 8 - Data from the Interviews • Number of women interviewed in March — April 2000 20 • Number of women interviewed in May — June 2000 a 12 • Total number of interviews conducted 32 a Note: The women who were involved in the second round of interviews were all interviewed in the first round in March 2000. Participant Observation and Fieldnotes Observation has been characterized as "the fundamental base of all research methods" in the social sciences (Adler and Adler, 1994, p.389) and as the "mainstay of the ethnographic enterprise" (Werner and Schoepfle, 1987, p. 257; cited in Angrosino and Mays de Perez, 2000). According to Adler and Adler (1994), observational research is a powerful source of validation. Observation rests on something researchers can find constant — their own direct knowledge and their own judgement. The keeping of careful and detailed fieldnotes over time is meant to serve as a protection against the inevitable loss of sharpness of perception as the researcher becomes a more central member of the community under study (Acker, 2000). In my role as the Project Manager I was a participant observer at most W O A W meetings and recorded fieldnotes; I wrote fieldnotes for over 50 subgroup and Project Team meetings. My involvement as a participant observer in Phase 0 (September to December 1999) helped me hone my research questions, determine appropriate language, and gain an intuitive understanding of WOAW. I continued recording fieldnotes throughout my dissertation, during and after the Research Team meetings, the interviews, W O A W meetings (Project Team meetings, W O A W subgroup meetings), phone conversations, informal discussions, and email correspondences. I also kept a journal to document my personal reactions, assumptions, feelings, and insights. For many of the Research Team meetings there are two sets of fieldnotes since another member from the UBC Working Group attended. In table 9, a breakdown of the fieldnotes that were recorded is provided. The guidelines that were followed for the participant observations are in 84 Appendix G. Table 9 - Data from the Fieldnotes • R T M fieldnotes written by Colleen 18 • R T M fieldnotes written by other UBC researchers (Beth Pmnington, Pam Ponic, Sydney Millar) 11 • Ongoing reflective journal ongoing • Total number of sets of fieldnotes 29 Data Analysis: "Sense Making" The Analysis Process Denzin (1994) refers to the data analysis process as "sense making" - how the writer moves from data gathering to the actual writing process, making decisions about what will be written about, what will be included, how it will be represented, and so on (Denzin, 1994). I engaged in an analysis process similar to many quaUtative researchers. I used Adas.ti 4.1 to organize, sort, and code my data; the ways that my use of Adas.ti enhanced the trustworthiness of the analysis are explained in the section "Legitimation and Tmstwortiiiness." Coding is the heart and soul of whole-text analysis, and forces the researcher to make judgements about the meanings of blocks of text. The fundamental tasks of coding include sampling, identifying themes, bunding code-books, marking texts, constructing concepts, models, and theory, and testing these models against empirical data (Miles and Huberman, 1994; Ryan and Bernard, 2000). My data set included 71 primary documents; refer to table 10 for an overview of my data Table 10 - Overview: Dissertation Data • Number of interview transcripts 32 • Number of Research Team meeting transcripts 10 • Number of sets of fieldnotes 29 • Total number of primary documents for analysis 71 85 The analysis process I adopted is best described by Ryan and Bernard (2000). I used chunks of text as the unit of analysis, and they were assigned "codes" representing thematic units. I began my analysis with general codes that I derived from reading the hterature and my work as the Project Manager. The coding was an iterative process; I went through the data several times, recalled codes, and reconfirmed my analysis. Through this process I revised and added codes and subcodes. I developed a "code book" to organize the list of codes; it appears in Appendix H . The coding process involved marking quotations from the primary documents with code names. The marking of text enabled me to see which codes reoccurred, to determine the codes that required further examination and the codes that were too specific. I also used the Adas.ti function "autocoding" to search all of the primary documents for key words such as "poverty," "disability," "health," and "stress," and to double-check the consistency and thoroughness of my coding. I did some counting with the coding and autocoding to assess the weight of the emerging themes. For instance, the code "poverty-barrier" had 120 quotations and required a more detailed analysis. I recoded the "poverty-barrier" code to more specifically capture the reason and nature of the barrier — was it a consequence of having little money, the treatment of welfare workers, or something else? Through this process I recorded code and theory memos and operational notes concerning emerging conceptual and theoretical understandings. Memoing is a principle technique for recording relationships among themes (Denzin, 1994). Finally, through Adas.ti's "network" function I visually represented the relationships between concepts and analytic terms. The Realities of Collaborative Data Collection and Analysis Feminist action research is characterized by the negotiation of description and interpretation (Lather, 1991). The early and final stages of my dissertation were not fully collaborative. I 8 6 established my research questions prior to meeting the women in W O A W or the Research Team. I also wrote the interview questions and the questions I posed at the first three Research Team meetings. Before the formation of the Research Team in March 2000, my involvement with W O A W enabled me to confirm that my questions were relevant and important to the women. Yet I was in the position to detemiine how the Research Team and interview discussions transpired and I had some influence on the major issues that arose. Similarly, I also r influenced who remained involved with the Research Team and who did not. The women who gravitated to my questions and felt that they were meaningful to them remained involved. Possibly, the women who did not stay involved with the Research Team felt less attached to or comfortable with the questions I asked. The data analysis process for this dissertation was iterative, dynamic, and partially collaborative. At seven of the Research Team meetings (April 24, 2000, May 16, 2000, June 7, 2000, June 29, 2000, August 17, 2000, September 27, 2000, April 3, 2001, May 29, 2001) I either distributed summaries and themes from the previous meeting's discussion or wrote major discussion topics and outcomes on a flip chart. I was diligent about transcribing the meeting widiin a week so that I could extract the major discussion topics, decisions, outcomes, and themes, and present them at the following meeting. Early on I had determined that this was an ideal way to collaboratively analyze the data while furthering the conversation. After the third Research Team meeting I wrote: I then put up the flip chart with their descriptions of "the label of welfare" and said that this is what I learned from the last meeting. The women were really interested in this and a few asked if they could get copies. I said that at the next meeting we could go over what they had said, and what I had gleaned from the conversations, and add to or amend their comments. They were all very interested in this and I realized that this may be a great way of guiding the discussion (Colleen's fieldnotes April 26, 2000). Freire referred to this process as "starling where the people are at" through fmding the "generative themes." Generative themes are those issues that spark animated discussion among 87 people because they are "close to the heart" (Freire, 1996). As we collaboratively analyzed the data I probed and questioned the women with open questions and invited reflection. Susan said that she appreciated the opportunity to collaboratively analyze: I appreciate the fact that we're working together to see how we're working together as a group. 'Cause I've done research before where you're just "the person," and you never get the feedback. Actually, sometimes you don't even get the feedback! You just give your input and that's it. It's nice to know how we feel is really important, 'cause we're on the opposite end of the table (Susan, R T M April 4, 2000). The most in-depth collaborative analysis occurred at the Data Analysis Potiuck in May 2001.17 At this meeting I went through my code-book. I framed the discussion as "things I had learned," and with each theme asked if I had "got it right and if anything was missing." There were a few additions and revisions to my findings, but the women said they were comfortable with my analysis. Despite my attempts at collaborative data analysis, I cannot claim that the analysis was fully collaborative. Certainly I did "member checks" throughout the data collection, but after the Data Analysis Potiuck I continued with the analysis through writing my dissertation and had litde input from the women. The later stages of analysis and writing are the most difficult for collaboration, particularly for graduate students with timelines. In this regard, I feel that I did not meet the ideals of FAR. Although feminist action researchers contend that it is possible and desirable to collaborate with research participants through all stages of the research process, a growing number of critics suggest that FAR is partisan, partial, incomplete, and inextricably bound to the contexts and rationales of the researcher and the positions she may knowingly or unknowingly represent (Regehr, 2000). The use of FAR strategies does not ensure that rapport and trust are built, or that research participants are fully comfortable with the researcher and with honesdy portraying their lived realities. In feminist action researchers' determination to have women tell their stories, 88 at times we have forgotten that stories are told to particular audiences and that sometimes women do not want to tell their stories. Research participants may feel forced to reveal personal information, may tell us what they think we want to hear, or may fabricate stories. Indeed, research participants construct narratives for specific purposes and audiences (Poland, 1998), and honest and faithful accounts may be contradictory or inconsistent (Alvesson and Skoldberg, 2000). In conversation people are constandy positioning themselves, positioning others, or being positioned by others as powerful or weak, cautious or bold, virtuous or evil, and so on, drawing upon culturally shared notions of what it means to be these things (Tierney, 2000). Importandy, no text is singularly created. It is not a matter of simply uncovering truths and untruths, but of exploring the role of the author in translating the text. For sure, the questions asked can frame how interviewees think about what to say (Tierney, 2000), and meanings are not unitary or fixed but multiple and shifting and depend on who is doing the interpreting and in which context they are used (Drew and Dobson, 1999). Despite FAR's inclusive and participatory ideals and my efforts to collaborate at most stages of the research process, it was impossible to escape my inextricable role as the feminist action researcher in a position of power and privilege to transform the women's words into this analysis. Power and Representation: Attempts at Deliberate Reflexivity With the postmodern age many qualitative researchers assert that we are in a crisis of representation. Lincoln and Denzin (2000) claim that this crisis speaks to the representation of the Other in our texts and the authority we claim for our texts. They ask: "Who is the Other?"; "Can we ever hope to speak authentically of the experience of the Other, or an Other?"; and "How do we create a social science that includes the Other?" (Lincoln and Denzin, 2000). No longer is the researcher an objective observer or data collector; rather, the researcher must be conscious of her location and subjectivity in the narratives constructed. What knowledge we are 89 able to observe and reveal is direcdy related to our vantage point, to where we stand in the world. Our interaction with the social world is affected by such variables as gender, race, class, income, sexual orientation, age, and physical ability. This does not mean that facts about the world do not exist, but that what we see and how we go about constoicting meaning is a matter of interpretation (Kirby and McKenna, 1989). Since I am a white, middle-class, educated woman who has never lived in poverty nor had children, it was particularly important for me, as a feminist action researcher, to reflexively consider the power differences between myself and my research participants. Researchers who adopt a reflexive stance do not see themselves as occupying a privileged position outside the world they study. The research they engage in is not a neutral procedure for discovering an "objective" external reality that exists independent of human perception and interpretation. The aspects of the environment that are noticed and singled out for inquiry, and the procedures which are used to describe and explain phenomena are ideological in the sense that they are socially constructed in a particular time and place (Poland, 1998). In fact, some researchers contend that the Other, who is presented in the text, is always a version of the researcher's self. Writers are positioned outside, yet alongside, those Others they write about, never making clear where they stand in these hyphenated relationships that connect the Other to them (Denzin, 1994; Fine, 1994). At its core reflexivity is about reflecting on power — a researcher's power to perceive, interpret, and communicate about Others. In her discussion of developing authentic student-faculty relationships, Tom (1997) writes about being "deliberate." A "deliberate relationship" is one that is entered consciously and ethically. Often the reasons for wanting and entering a relationship are unclear and very different - "because I have initiated the relationship, they may not be as clear about what they want as I am about what I want" (p.4). As a feminist action 90 researcher I strove to be "deliberately reflexive," to take responsibility for thoughtful and reflective practice and to remain conscious of the inherent power dynamics in the research process. In one interview, when I asked Elizabeth about the benefits and challenges of partnering with researchers, she questioned whether researchers were able to understand and represent her reality. Colleen: What are the advantages and challenges of partnering with researchers? Elizabeth: Well, the challenges may be not knowing the area or what our obstacles are. Or maybe not being, I don't know. Colleen: Not being.. .not being the same? Elizabeth: Yeah, like, I don't know people's financial situations. It's like, how well do you really understand the low income thing and being in the same boat, you know, that aspect. Colleen: So do you feel like in order to be able to represent what's going on, being on low income and being a single mother, that you have to be living on low income and be a single mother? Elizabeth: Maybe not at present, but it certainly couldn't hurt to have had it at some point so you can understand where we're corning from. Colleen: So how do you feel, let's say for the next set of interviews, there's me asking you about your experiences about being a single mother and living on low income. How does that make you... are you comfortable with that if you know that I'm not in the same situation as you? Elizabeth: Well, oh yeah. That's ok. Just some people you've come across it's like "I'm better that you" and snooty. You know, you're very easy to get along with (Elizabeth's interview, March 20, 2000). Elizabeth said that she was "ok" with me interviewing her, yet her comments highhght the need to be consistendy and deliberately reflexive throughout the research process. Indeed, being deliberately reflexive does not resolve "power plays," but it can increase the complexity of the research process by employing an analytical approach that doubles back on itself (Ristock and Pennell, 1996). As well, in my efforts to be deliberately reflexive I attempted to be as vulnerable as possible. As Richardson (2000) suggests, I worked to be self-aware and self-exposed and to hold myself accountable for the standards of knowing and writing about women I studied (Richardson, 2000). In my reflective fieldnotes, memos, and research journal I documented my assumptions, 91 my uses of power, and my reflections on the process of knowledge generation. My purpose in being deliberately reflexive was to improve the quality of my research and to stay accountable to the research participants through being as authentic, truthful, and vulnerable as possible. A text that displays honesty and authenticity "comes clean" about its own stance and about the position of the author (Lincoln, 1995, p.4). A reflexive researcher ask the reader to accept her text as authentic, that is, as a conscientious effort to be truthful about the making of the account (Gergen and Gergen, 2000). Through all stages of my research project, and particularly when considering my withdrawal from the field, honoraria payments, and anonymity and confidentiality, I strove to be deliberately reflexive. Withdrawal From the Field It was important for me to make deliberate choices about my behaviour and the long term goals for the relationship I developed with the research participants (Tom, 1997). As a feminist action researcher committed to being deliberately reflexive I decided to not withdraw from the field at the completion of my data collection. Withdrawal from the field in order to analyze and write is a typical phase in most research projects, particularly for graduate students who are confined by university timelines and limited resources. I initiated the Research Team and experienced its development into a meaningful place for the women to share experiences and explore ideas for action. I was committed to the FAR process and to my role on the Research Team, and withdrawing because I had completed my data collection felt artificial and unethical. Despite the differences between myself and the women on low income, I was not only a researcher, I was also a Research Team and W O A W member. Although members of my research committee encouraged me to withdraw completely, I remained involved but shifted the nature of my involvement. I continued to record fieldnotes and to tape record the meetings but did not transcribe or provide summaries of them.18 Indeed being deliberately reflexive made me 92 question mainstream assumptions that it is possible and desirable to withdraw from the field when the data collection is completed.19 Honoraria Payments Through the SSHRC research grant the women were paid for their involvement in the interviews, in the first 4 Research Team meetings, and in the Research Parties.20 Given the monetary value of the research grant, I felt that the women should financially benefit from their participation — it would have seemed inherendy inequitable if the women did not receive some compensation for their involvement. Both the researchers, through the SSHRC grant, and the service providers, through their employment salaries, were paid for their involvement in WOAW. Yet I was conflicted about paying the women. I was concerned that the honoraria would coerce some women into participating on the Research Team. I hoped that the Research Team members would participate because of their interest in the discussions and the potential for action, not because of an honorarium. I raised this concern at several Research Team meetings and the women acknowledged that some may be motivated to attend the Research Team meetings because of the honoraria. They also suggested that they appreciated and looked forward to receiving the honoraria payments. The Working Group decided that from Phase 2 onwards the honoraria would be attached to research activities but not to the Research Team. Some women did not return to the meetings, though a significant core of women remained involved. By virtue of my job as Project Manager to manage the SSHRC research grant's budget I retained control and power over the women through being in the position to determine how, when, and for what they got paid. I was constandy conflicted about the honoraria and found no resolution. Even though honoraria payments are a difficult issue laced with power and ethical dilemmas, as a deliberately reflexive researcher I had to confront these challenges and make the best choices possible. 93 Anonymity and Confidentiality Anonymity and confidentiality were discussed at the initial Research Team meetings and in the interviews. The women were aware of the openness of the Research Team and I reinforced the fact that it was a public forum - any W O A W woman could attend, at times some women invited friends, and the events and outcomes of meetings were not confidential. To avoid the circulation of raw data and to prevent the potentially harmful identification of the women I did not distribute Research Team meeting transcripts. When a participant wanted to review a transcript, we did so together and I retained the copy of the transcript. O n several occasions I asked the women about using pseudonyms, and each time they said they wanted to use their real name.21 Yet throughout the research process I was concerned some women could be harmfully identified, particularly since several were critical of the welfare system and their workers. In May 2002 I mailed each woman all of her quotations that I used in my dissertation. In detail I outlined the possibility that they could be identified if they chose to use their real names, and that this could have negative repercussions for them. Consequentiy, in follow-up phone conversations, only 4 women wanted to use their real (first) names; all but 2 of the women who preferred a pseudonym chose their own. Al l of the researchers and the community service provider were comfortable with using both their first and last names, possibly because they were not the focus of this study. Regardless, the fact that most of the women on low income chose pseudonyms and that none of the researchers or community service providers did highlights the power differences between these groups of participants and how engaging in feminist action may carry different and possibly more severe risks for women on low income. This is further explored in chapter 7. Legitimation and Trustworthiness The objective truth about a society or a culture cannot be established (Angrosino and Mays 94 de Perez, 2000). In fact, "in the social sciences there is only interpretation. Nothing speaks for itself (Denzin, 1994, p.500). Given that objectivity cannot be established and that there is only interpretation, qualitative researchers have long asked how a text legitirnates itself, or makes claims for its own authority. Seidman (1991) proposes a local, personal, and pohtical turn in the postmodern era. "Instead of appealing to absolutist justifications, instead of constructing theoretical logics... I propose that we be satisfied with local, pragmatic rationales for our conceptual approaches" (Seidman, 1991, p.136; cited in Denzin, 1994). The criteria for legitimating the trustworthiness of my data analysis were crystallization, credibility, and transferabihty. Adas.ti, a qualitative data analysis software program, was a tool that facilitated a more legitimate analysis. Crystallization Crystallization recognizes the many facets of any given approach to the social world as a fact of life fjanesick, 2000). According to Richardson (1994), the crystal "combines symmetry and substance with an mfinite variety of shapes, substances, transmutations, multidimensionalities, and angles of approach. Crystals grow, change, and alter, but are not amorphous— Crystallization provides us with a deepened, complex, thoroughly partial, understanding of the topic. Paradoxically, we know more and doubt what we know" (Richardson, 1994, p.522). There were two ways that I addressed crystallization as a tmstworthiness criteria. First, the different data sources - the Research Team meetings, one-on-one interviews, and fieldnotes from multiple authors — provided different perspectives, or angles of approach, to the research questions. Second, multiple lenses were used in the analysis and writing. As described in chapter 1,1 approached chapters 4, 5, 6, and 7 with different voices, literatures, and analytic perspectives in an attempt to represent different facets of the data. Both strategies provided a more complete picture, meanwhile raising contradictions. Crystallization allowed me to weigh the evidence 95 because "slices of data" were drawn from different data sources, researchers, and analytic perspectives. Credibility Qualitative researchers do not claim that there is only one way of mterpreting an event. The traditional assumption that the truth can be established through careful cross-checking of researchers' and insiders' reports is no longer universally granted. Contemporary social scientists are increasingly mclined to expect differences in testimony grounded in gender, class, ethnicity, and other factors (Angrosino and Mays de Perez, 2000). Indeed, there is no one "correct" interpretation (Janesick, 2000). Validity in qualitative research has to do with the credibility of the description and explanation and whether or not the explanation fits the description. According to Greenwood and Levin (1998), credibility is "the arguments and the processes necessary for having someone trust the research results" (Greenwood and Levin, 1998, p.80). There are three types of credibiHty - internal, external, and catalytic. Internal credibihty is the degree to which the research participants agree with the research findings. The process of collaborative data collection and analysis rendered this account internally credible since the research participants were involved in the construction of meaning. The process of sending quotations back to the research participants after the dissertation was written also contributed to the internal credibihty of the analysis.22 I conducted "member checks" with the UBC Working Group and members of the W O A W organization. At various times different UBC Working Group members attended and recorded fieldnotes at Research Team meetings. At many Working Group meetings we discussed the Research Team and I was able to check my assumptions, experiences, and emerging analyses with them. As well, I co-wrote the Phase 1 and 2 W O A W Research Reports (Reid, Frisby, 96 Millar, and Pmnington, 2000; Reid, Frisby, Millar, Pmnington, and Ponic, 2001). These reports summarized the research activities and provided some initial research findings and reflections. In the Phase 1 Report I included a section on the women's health issues and the relationship between health and poverty, and in the Phase 2 Report an analysis of the Research Team meetings was provided. After the reports were distributed, the Working Group met with W O A W members to check the credibihty of our analyses. We asked for additions, deletions, and topic suggestions for upcoming reports. Working with the UBC Working Group and co-writing and vakdating the Phase 1 and 2 Reports with all W O A W members contributed to the external credibihty of my fmdings. The Working Group and the broader W O A W organization were not direcdy involved in my study though they were familiar with it and the women involved said that the fmdings were convincing and credible (Greenwood and Levin, 1998). Catalytic validity represents the degree to which the research process reorients, focuses, and energizes participants toward knowing reality in order to transform it. Catalytic vahdity lies not only in recognizing the reality-altering impact of the research process, but also in the desire to consciously channel this impact so that respondents gain self-understanding and, ultimately, self-determination through research participation. Over the course of the Research Team meetings the women identified shared problems, barriers, and desired actions. Despite its challenges (addressed in chapter 7), working together towards social change increased the catalytic credibihty, or "workability," of the actions, and enhanced the women's sense of control over their situations (Greenwood and Levin, 1998). Transferability FAR-developed knowledge can be valuable in contexts other than those where it is developed. Transferring knowledge from one context to another relies on understanding the contextual factors of the situation where the mquiry took place, judging the new context where 97 the knowledge is supposed to be applied, and making a critical assessment of whether the two contexts have sufficient processes in common to make it worthwhile to link them (Greenwood and Levin, 1998). In terms of this study, important considerations are the demographics of the research participants, the structure of W O A W and the Research Team, the purpose, configuration, and length of the SSHRC research project and my dissertation, and the varying involvements of the service providers and researchers. Other important considerations are the pohtical, social, economic, and cultural environments in which this study occurred. Atlas.ti 4.1: A Trustworthiness Tool To manage the volume of data gathered and to ensure the comprehensiveness of my analysis I used the qualitative data analysis software program Adas.ti 4.1. Software can help competent researchers do more rigorous, consistent, and thorough analyses than they otherwise might. It is not that the software itself makes it more rigorous, but that the researcher can use the software to do more rigorous work than she could do without it (Weitzman, 2000). According to Weitzman (2000), using Adas.ti can increase the rigor of an analysis in four ways - through faciUtating consistency, speed and efficiency, representation, and consolidation. Adas.ti enabled me to search all the places a given key word, code, and combination of codes appeared. I was able to review all the data assigned to a particular code and check my initial interpretations. With Adas.ti I searched and re-searched almost instantaneously to find data that applied to a particular question. I quickly resorted my database, redefined codes, and reassigned chunks of text. The speed and efficiency afforded by Adas.ti encouraged me to check and revise my analysis whenever necessary. Additionally, Atlas.ti facilitated the dynamic representation of my thinking through identifying themes, building codebooks, marking texts, creating memos, and developing theoretical models (Ryan and Bernard, 2000; Weitzman, 2000). Finally, with I kept everything in one place. The consolidation of my data freed up large amounts of energy for me 98 to see and keep track of connections that might otherwise easily fall through the cracks (Weitzman, 2000). Writing Just as conducting interviews and recording fieldnotes are essential skills for the qualitative researcher, so is writing. Al l too often our writing suppresses individual voices and accepts the omniscient voice of science as if it were our own. Richardson (2000) contends that writing is a method of inquiry and should not be taken for granted — it is a way of knowing and a method of discovery and analysis. Writing is how we "word the world" into existence (Rose, 1992; cited in Richardson, 2000). Although this "worded world" never accurately, precisely, or completely captures the studied world, approaching writing as a method of mquiry honours and encourages our attempts and provides a research practice through which we can investigate how we construct the world, ourselves, and others (Richardson, 2000). Qualitative writers can challenge the questionable meta-narratives of scientific objectivity and still have plenty to say through engaging our subjectivities in the knowing and telling about the world as we perceive it (Richardson, 2000). A postmodernist position does allow us to know "something" without claiming to know everything. Having a partial, local, historical knowledge is still knowing, and this perspective can free us from trying to write a single text in which we say everything at once to everyone (Richardson, 2000). As we write we make decisions about what will be written, and as one moves from a blank page to a written text, an emergent, reflexive interpretation of the subject matter is built (Denzin, 1994). Two core principles of feminist action research are researcher reflexivity and collaboration through all stages of research. I attempted to be deliberately reflexive and transparent throughout the data collection, analysis, and writing, particularly in terms of my own assumptions, interactions, analyses, and relationships with the research participants. In a feminist 99 action research project all participants should have a role in negotiating the final meanings of the research (Lather, 1991) and in writing the report. There has been far less feminist attention paid to the processes of retairiing the research participants' voices in the phases of data analysis and writing up (Edwards and Ribbens, 1998). I did a significant amount of analysis through the writing process, and while I did member checks with all W O A W members to check the credibihty my analyses, this process was not fully collaborative. Although writing may be one of the most difficult aspects of a research project for collaboration, collaborative writing is a strategy for increasing the internal vaUdity and authoritative claims of the text. In the future I hope to explore the potential for collaborative writing in producing authentic texts from the research participants' perspectives that would be consistent with a feminist action research framework. While qualitative researchers are in a crisis of representation, legitimation, and writing (Lincoln and Denzin, 2000), the development of research methods consistent with a feminist action research framework can illuminate these crises and provide strategies for addressing them. In particular, FAR's principles of inclusion, participation, and researcher reflexivity are a starting point for increasing the transparency and credibility of our work. The FAR context and methods that were described in this chapter both upheld FAR's ideals and reflected more traditional qualitative approaches. Nevertheless, the continued development of FAR methods can advance understandings of the quahtative crises of representation, legitimation, and writing. Notes 1 A t that time shared identity was the basis for two of the subgroups — Aged to Perfection (senior or older women) and Iranian Mutual Support Group (recent immigrant women from Iran). The other two subgroups were based on the women's geographic communities - P o C o W O A W (Port Coquidam) and S W C o W O A W (South West Coquidam). It so happened that the P o C o W O A W subgroup was primarily single mothers, while the S W C o W O A W subgroup had a more diverse group of women involved. 2 The number of service providers involved with W O A W remains approximately the same, though some have discontinued their involvement and newer partners have more recendy joined W O A W . The service providers 100 represent different organizations in the local communities, including community parks and recreation (6), a woman's centre (1), local social service agencies (2), and a community school (1). 3 CoPoMo W O A W is short for the Coquidam Port Moody W O A W subgroup. 4 The reasons why the group of primarily immigrant women (Iranian) disbanded are complex and poorly understood. Possible reasons for the dissolution of this group are presented in chapter 4. 5 Wendy secured a seed grant from SSHRC ($5000) in July 1999. The purpose of the SSHRC seed grant was to support the writing of the larger research proposal to the "Society, Culture, and Health of Canadians" SSHRC competition. I helped Wendy with formatting, revisions, the letters of support, and compiling the curriculum vitae of the supporting research collaborators. 6 The name of the project on the application to SSHRC was "Addressing the Self-Determined Health Problems of Low-Income Women Through Participatory Community Interventions: Meanings, Process, and Evaluation Issues." The project was rarely referred to by this title, and was more commonly known as W O A W "Women Organizing Activities for Women." 7 In her Master's thesis, Sydney Millar, one of the UBC Working Group members, examines the "community development" discourse that was used by three of the W O A W service providers. She also examines the context in which this discourse was used (community recreation, social services, advocacy organizations). For a more in-depth analysis of community development discourses consult Millar (2002). 8 There are substantial literatures on both "collective organizing" and "community development." W O A W identified as a collective organization as a consequence of making decisions through consensus and promoting a flat (versus hierarchical) organizational structure. It is beyond the realm of this dissertation, however, to provide a more in-depth analysis of them. 9 For the purposes of the research study I established a data management and analysis system (Atlas.ti 4.1), a budget management system (Excel 2000), a research bibliography (Endnote 4.0), a database to track project attendance and activities (Access 2000), and an office management system for tracking correspondences and events. 1 0 A group of interdisciplinary research collaborators, who had expertise in different areas that related to the SSHRC grant, were distantly involved in the project. Aside from Dr. Allison Tom (UBC), the collaborators were not actively involved with WOAW. Allison Tom was also my research supervisor and had regular interactions with the UBC Working Group. The other collaborators, from UBC and from SFU, were Drs. Heather McKay, Susan Crawford, Bonnie Long, and Peter Crocker. 1 1 The honoraria were problematic throughout the research project. The ethical dilemmas regarding the honoraria are explored later in this chapter. 1 2 Note that not all Research Team members were research participants. See section "The Women's Involvement as Research Participants" for an explanation of their involvement in the dissertation. 1 3 The structure of the W O A W organization and how I believe it facilitated the women's continued involvement in the Research Team is further explored in chapter 7. 1 4 Although the Research Team was initiated for the purposes of my dissertation and was initially an unanticipated aspect of the SSHRC project, all data that were gathered by researchers working with the SSHRC project belong to it. However, in order for any other UBC Working Group member or research collaborator to use data from the Research Team meetings, agreement must be reached between myself and that person. 1 5 These women were interviewed by Pam Ponic for the purposes of the larger SSHRC research project in June 2001 with Interview Schedule #1 (see Appendix E). 1 6 The 71 primary documents totaled over 1600 pages of data. 1 7 In April 2001 I asked the women if we could devote an entire Research Team meeting to collaboratively analyzing and reviewing the findings for my dissertation. They agreed that they did not want materials to read in advance but would prefer if I did a presentation (some joked that they expected power point slides with the presentation). They also suggested that the meeting be a poduck and extended to 3 hours. It then became known as the "Data Analysis Potiuck." 1 0 1 1 8 This refers to distinction I made previously between primary transcripts and secondary transcripts. The secondary transcripts (meetings from January 2001 onwards) were transcribed by a UBC Working Group member and have not been analyzed to date. At the very least these transcripts are a good record of the evolution of the Research Team, though in the future I plan to use the data from these meetings for analysis and writing. 1 9 As discussed in the epilogue, my ongoing involvement with the Research Team (beyond the primary data collection phase) did not challenge this analysis, rather it deepened my understanding of the complexities of engaging in feminist action research and the difficulties faced by many women on low income to particpate in such endeavours. 2 0 The women were paid $100 for being interviewed and participating in 2 Research Team meetings. They were paid an additional $100 for being interviewed a second time and participating in 2 more Research Team meetings. After Phase 1 of the SSHRC project, the honoraria was no longer based on attendance at Research Team meetings. 2 1 The consent forms said that the women could choose to use their real names or a pseudonym. See Appendix C. 2 2 Each participant received an outline of my dissertation with all of their quotations that I used. I called each participant and asked if they were comfortable with what they read. Al l of them reflected that my analysis was accurate and captured our Research Team and interview discussions, and only one woman asked that I clarify something in a footnote (in chapter 5). 102 The Women's Stories Multiple small spheres of personal experience both echo and enable events shared more widely, expressions of moments in which we recognize that no microcosm is completely separate, no tide pool, no forest, no family, no nation. Indeed, the knowledge drawn from the life of some single organism or community or from the mtimate experience of an individual may prove to be relevant to decisions that affect the health of a city and the peace of the world. Mary Catherine Bateson In this chapter I attempt to provide a sense of the women's complex hves and experiences. I tell the stories of the 12 women who were twice interviewed and who were actively involved in W O A W and the Research Team in the year 2000. Since I did not embark on my dissertation with the intention of gathering life stories, the subsequent stories have gaps and omissions. Through the lens of my research questions, I formulated partial stories based on crude classifications. As well, the stories are not equal in length — half are long and detailed because of availability of data, while the others are significantly shorter. The stories are grouped according to the women's W O A W subgroup affiliation. After the women's stories I have drawn together emergent themes that relate to the research questions I posed in this dissertation. 103 Table 11 - Profile of Individual Research Participants (n=20) Name Age Domestic Status WOAW subgroup & status Education Race/ Ethnicity Major health concerns Financial support Alexa 28 Single no children CoPoMo - active Grade 12 White Arthri t is, depression Disability benefits I Arlene > 60 Single (separated) 2 adult children Aged to Perfection - inactive Grade 12 White Fibromyalgia Seniors pension Caroline 36 Single (divorced) 3 children, aged 5, 8, and 11 PoCo - inactive Don ' t know White Clinical depression Social assistance Cassie 40 Married 2 children, aged 1 and 3 CoPoMo - active Bachelor o f Human Kinetics Chinese Depression, back problems, stress Work ing husband Cynthia Late 20s Single (divorced) 2 children, aged 2 and 5 PoCo - inactive Grade 12 White Depression, stress Social assistance Elizabeth 32 Single (separated), 1 child, aged 4 SWCo - active Grade 12 White Hyperthyroidism, eating disorders, post-treatment breast cancer Social assistance Gloria Late 40s Single (divorced) 2 adult children SWCo - inactive Don ' t know White Depression Ex-spousal support Joanne 30 Single (widowed) 2 children, aged 2 and 5 PoCo - inactive Grade 12 African American Hyperthyroidism, migraines Social assistance Julie 28 Single (divorced) 2 children, aged 2 and 4 PoCo - active Some high school White Learning disability Social assistance Katharine 50 Single (divorced) 2 adult children CoPoMo - active Grade 12, some university White (French Canadian) Mult iple sclerosis, clinical depression Disability benefits I Kelly 34 Single (separated) 1 child, aged 3 PoCo - active Grade 12 White Endometriosis, osteoporosis Social assistance Maey 33 Single 1 child, aged 10 SWCo - active Some elementary White Eating disorders, "elephantitis," scoliosis, learning disability (dyslexia) Disability benefits II Martha >60 Single no children Aged to Perfection - active Don ' t know White Chronic fatigue syndrome, fibromyalgia Disability benefits II Rene Late 50s Single (divorced) 2 adult children Aged to Perfection - active College diploma, art school White Angina, diabetes, irritable bowel syndrome, high blood pressure, sleep apnea Disability benefits II 104 Table 11 - Continued Name Age Domestic Status W O A W subgroup & status \ Education Race/ « Ethnicity Major health if. concerns Financial support Susan 44 Single (separated) no children SWCo - active College diploma White Fibromyalgia, diabetes, clinical depression, irritable bowel syndrome Disability Benefits I I Teresa mid -40s Married 2 children, aged 11 and 14 CoPoMo - active Don't know Chinese Diabetes, arthritis On medical leave; has a working husband Trina 56 Single (divorced) 3 adult children Aged to Perfection - active Some high school White Clinical depression, arthritis Disability benefits I I Virginia Dawn 25 Single (divorced) 1 child aged 3 PoCo - active Grade 12 White Depression Social assistance Wanda 61 Married (twice widowed) 7 adult children Aged to Perfection - active Grade 12 White Diabetes, sleep apnea, high blood pressure Seniors pension Willow 26 Single, 2 children, aged 1 and 4 SWCo - active Some high school White, First Nations children Social anxiety disorder, depression, stress Social assistance Note: These data were compiled from the 20 women who were interviewed in the first round of interviews. They appear in alphabetical order by first name. The Women's Lives and Experiences1 Aged to Perfection Subgroup Rene Rene, a single woman in her late 50s, had a diverse past. She had worked in bush camps, as a jade cutter, had gone to art school and computer school, been a Tupperware manager, and led a union in solidarity. When she was 16 she was made a ward of the Catholic Children's Aid after her father's suicide. I think a lot of my attitude and that comes from the fact that when I was sixteen my father killed himself. My mother was in hospital. And I was made a ward of the Catholic Children's Aid. I was on my own virtually. My sisters were married and had kids. And I was very independent, even before that because my mom was in hospital. And my dad and I were baching from the time I was thirteen. We were resourceful and independent right from the begmning. And I think a lot of the counseling in the Catholic Children's Aid gave me made me resourceful. I got a good worker, a really good worker at that time. So even then I knew how to ferret out what I wanted to know. And I was not a 105 passive person in any sense (Rene's interview, June 5, 2000). In her 20s she became a jade cutter, and then moved into the "bush" with her husband. I worked at that [cutting jade] 'til I got married and went to the islands. Silly girl... I was up the Alaska Highway. And I discovered that I was left alone in camp with my child. And it wasn't the four-legged animals I was afraid of. It was the ones that wandered in off the road. So I went and got a gun and learned how to use it.... We worked together in [bush] camps. He [husband] had one leg amputated. He wore an artificial limb. And I would work along side him rather than take the passive role. I would help with the carrying and so on and so forth. And put in the 18 hour days. And look after my kids. So that I think that has more to do with the health and the way I am today than anything else. And when I kicked him out I was looking working full time as a head cook (Rene's interview, June 5, 2000). After leaving her husband, Rene had a series of jobs and at one point chaired her union. I used to be a very shy person and then I started selling Avon door to door. I did it purposely to make myself get out and meet people 'cause I didn't like doing it, and that was one step. And then I went to work at the hospital and they brought in a union and we kicked the union out and got another union. I didn't want a union, but that was the rules they gave us, so I joined the union and I got involved. 'Cause I wanted to learn. And I ended up chairing the union, but they gave me a lot of education in pubhc speaking and solving problems. So I've had a lot of opportunities other women may not have had. And a chance to use it. And then Tupperware, you get a lot of sales training and what not and you know, just how to co-operate in work and that sort of stuff. So, I've been very fortunate (Rene's interview, March 15, 2000). Rene felt that the work she had done and her experiences with her husband contributed to her failing health. And then him [husband] screwing around and bringing home some nasty diseases and me having to go on all these great antibiotics which was not good for the bowel. When I got rid of him I got rid of the big problem (Rene's interview, June 5, 2000). When I asked her about her health, she explained "I'm as healthy as can be under the circumstances.... I have IBS [irritable bowel syndrome]" (Rene's interview, March 15, 2000). Later at a Research Team meeting, Rene explained her vision of good health. For me, it would be being able to eat when I go out without worrying that's it a minefield, because I'll have the runs or be sick for days afterwards, and to be able to walk up and down the stairs without pain. And to get around, and not to have to sleep with a machine, because all these things limit my ability to travel, and to do things that I want to do, and it is hard. If I go somewhere I take food with me (Rene, R T M April 26, 2000). 106 Rene was politically active through her involvement with WOAW. She initiated a bus riders' union, went to meetings at city hall, joined protests, and wrote letters to her local MLAs. She chaired the board of her housing co-operative and ensured that people on low income were able to move in. Her days were filled; she was active in her community and worked with many W O A W women, teaching Enghsh, quilting, and cooking, and providing support. She was an avid quilter, and quilted through every W O A W meeting I attended. I saw Rene as resourceful, sharp, and tough, with rich and diverse experiences. Trina Trina was a 56 year old single woman with three grown children. She grew up in Vancouver with an abusive mother who moved constantly — in one school semester Trina went to five different schools. At 16 she left school and was forced to marry a friend of her mother's. She explained what happened when her oldest son was 4 months old. I didn't raise my first son, [he's] 37. My mom and my ex-husband kidnapped him when he was 4 months old. I had already left [my husband], but my mom and my husband came and kidnapped him. So I didn't see him for ten years. So I didn't bond with him at all. So when I had my daughter, I ran, so my mother couldn't find me (Trim's interview, March 21,2000). With her daughter and youngest son, she moved to a housing co-operative in East Vancouver. I lived in a co-op at 36 th and Main and my daughter was five when we moved in there and she was 13 when we moved out.... my children go to school with all these kids. And there's nothing but drugs and alcohol and abuse and everything in those little communities. And it was the biggest mistakes, living in that community. But I didn't know that at the time. Because my kids had everything. We had a nice home, we had three meals on the table and everything else. But our neighbours and everybody else didn't. Prescription drugs, alcohol, drugs of any kind. If I could have lived anywhere else on the money I was making we would have (Trim's interview, March 21, 2000). As a low-income single parent raising two children, she worked as a waitress "I was working at Denny's. Raising two kids by myself (Trim's interview, June 5, 2000). She explained: But being a single parent was my problem financially. I never got any assistance and I didn't stay at one job long enough to get into a pension or anything like that, so you're 107 just left with [being paid] by the hour and then, at 56, with nothing (Trina's interview, March 21, 2000). She spoke of the toll her work had taken on her body. I've always been tiny. I've been a waitress so it's all physical. I haven't had any serious accidents. It's just heavy Hfting, you know those great big tubs of dishes and serving. I can't even carry two dishes anymore. I love hobbies; I can hardly even crochet. I just can't hold it [crochet needle] 'cause of my shoulders.... And from my hips to my shoulders I am just so sore all the time. It's arthritis and tendonitis (Trina's interview, March 21, 2000). Since being diagnosed with arthritis, Trina had been on disability II. She had limited contact with her children — she never sees her oldest son and she sees her daughter and youngest son occasionally. My daughter just had her daughter, 8 weeks old just about. But even before she had her child she never gave me the time at all. She worked nights and I worked days and I'm a morning person; she's a night person. So if I didn't go see her, I didn't see her. The younger generation doesn't like the older one (Trina's interview, March 21, 2000). Trina said that until her mother died 4 years ago, she would not go through Port Moody for fear of seeing her and would only go shopping at night. For five years she had been involved in therapy and support groups to come to terms with her abusive past and her clinical depression. She was isolated and felt invisible. I don't know if you noticed, but I have three TVs. My isolation is, this is it. But it doesn't matter where I go in my tiny apartment, I can see something. I can see the TV, I can hear the voices. Even out on the patio (Trina's interview, March 21, 2000). Trina said that one of the W O A W service providers had called her every week for four months about W O A W meetings. In February 2000 Trina summoned the courage and attended an A T P subgroup meeting. Despite bouts of depression, isolation, and her initial tentativeness, Trina became increasingly involved in W O A W over the three years. She established friendships with the Aged to Perfection members and spoke up at meetings. O n several occasions I heard her support WOAW's consensus decision-making process and say that she had a voice. She was critical of women who she felt did not support collective organizing, and once asked me 108 "where's her sisterhood?" Despite her troubled past and fragility, she developed a sense of women's strength and seemed stronger over the course of being involved with WOAW. I think women are very powerful and once we put our minds to anything, we can do it. And I think we can, I really think we can reach around the world. We really can.... We keep the family together.... we're the glue (Trim's interview, March 21, 2000). Wanda Wanda was a 61 year old married woman who was twice widowed and had seven children and 14 grandchildren. She grew up in Rexdale, Ontario, and spoke of raising her three younger brothers. I was living at home, raising three brothers. I got up in the morning and did botdes and diapers and put them on the line. And came home at noon and took them off the line and went back after supper and fed them and cleaned them. I told you my mother did nothing.... And my husband, he was the oldest of five. And so was I, the oldest of five. And he had a business. His father was the town drunk. So it was Jack's responsibihty even if he wasn't working the shift to go back at night, close up, take the money home to his mother's place. And then open up again in the morning. Then we decided we were both tied down, why not getting married? We always figured that by the time we were 35 and 40 our kids would be grown and we could have our life then. But then husband died at 32. Yes, how great plans go astray (Wanda's interview, June 5, 2000). Wanda got married to her first husband just before her 17* birthday. I was married at seventeen.... I got married on the 7 t h of December. And my birthday's the 29 th. And I turned seventeen on the 29th.... I'd been dating my husband since I was eleven. And I was engaged for a year before I got married (Wanda's interview, June 5, 2000). Wanda and her first husband had five children and hved on a farm in Ontario. Although they did not have much money, she felt that her family ate well. My kids ate a bit better because we hved on a farm for the first ten years I guess of Bruce's life. I had a big vegetable garden. And we ate a lot of vegetables. Meat was expensive, vegetables were cheap (Wanda's interview, June 5, 2000). When her husband committed suicide at 32, she was left as a single mother and moved to British Columbia to be closer to her sister. After my first husband died I brought five kids and came out here [British Columbia].... 109 In '68. Somewhere around there. My husband died at 32. And I immediately reverted to being a child as far as my parents and his parents were concerned. And I thought I can't handle this. I've been on my own too long. So I packed the kids up. My sister was out here at the time. And I came out here with her. And then about two, three years later she went back. And I stayed (Wanda's interview, June 5, 2000). Within two years of hving in British Columbia Wanda remarried and had twin daughters. I had 7 [children], first 5 were all 15 months apart. And then I had a pair of twins, and I'd have twins any day, because they keep each other amused, they're going through the same stages at the same time (Wanda, R T M June 29, 2000). Wanda raised her seven children, had four foster children, and also had a career with Tupperware. She rarely spoke of being "poor" — rather, she suggested that she raised her children with limited finances. "I raised 7 children and I never had money, my husband worked, sure, but he was handicapped so we just didn't have it" (Wanda, R T M April 4, 2000). In her second interview she said: "They [children] learned what a second-hand store was. But I don't think that hurts anybody" (Wanda's interview, June 5, 2000). Wanda suggested that she needed to work to support her family, but that because she had a career her husband had learned to be a father. I worked for myself for 20 years, part of the reason I worked was to be away from the kids. I was raising 7 of them and I had 4 fosters at one time in the house as well. And it still worked. But the other tiling, that's the only time my husband became a father. If I had stayed home all the time, he never would have. He was quite happy not being a father (Wanda, R T M April 26, 2000). Through her career with Tupperware she had also made friends who supported her when her second husband died 10 years ago. The best thing that ever happened to both of us was when we became Tupperware ladies. As silly as that sounds.... When my husband died, those girls [from Tupperware] were through my door, and that was the only thing that kept me sane. They brought meals, they cleaned house, they looked after whatever needed to be looked after, and they were there to talk to me (Wanda, R T M April 4, 2000). When I met Wanda she hved in New Westminster with her trird husband. They were both pensioners; on several occasions Wanda said that she hved on $730 a month. She was active in 110 the local Shriners' Lodge, seniors' housing projects, and with her grandchildren. "I'm the recorder at the shrine That's the secretary-treasurer's job. And I do have time for my grandkids because I make time'' (Wanda's interview, June 5, 2000). Wanda spoke of her health as being the "mirror image" of her mother's. Most often, she referred to diabetes as her biggest health problem. My health very is almost like a mirror image of my mothers. Al l my mother's problems I have. My sister doesn't. Which is fortunate for her but we've discussed this. I have the legs and the feet like my grandmother. And my mother had. I have the high blood pressure. The diabetes. The whole nine yards. And my sister's almost as healthy as a horse. She's way overweight, but she's healthy (Wanda's interview, June 5, 2000). She also spoke of women's roles and emotional health. I think your whole health is tied up with your emotional health and if it's not right, nothing is going to go right. And particularly with women they fall in to this category very easy: "I don't feel well today, I think I have a headache." You know if they could get through the door and go do something their headache would be gone, but they have no place to go and nothing to do. So they stay home with their headache. We have some women who shop till they drop; that's never been my thing. They seem to get pleasure out of that. But if you haven't any money you can't even do that. So we end up at our age, and I'm really focused on my age right now because it seems to me that a lot of women are still the caregivers. They're still making the meals, they're still cleaning the house. A husband retires, but a woman never does. So all of a sudden she's sitting in the evening and what will she do? She'll watch the boob tube because she's too tired to go anywhere. Women get tired very easy. What used to take me an hour to do takes me sometimes all morning and sometimes longer (Wanda's interview, March 15, 2000). Wanda was an energetic and consistent W O A W member and often provided interesting insights into women's roles and responsibilities. In one case she said "The world's run by men, what can you say. We haven't changed that yet" (Wanda, R T M June 7, 2000). She adopted problem-solving roles for several W O A W women and was active in most meetings and activities. She believed in women's strength and capacity, and told me that she was increasingly interested in older women since she felt they had a lot to offer. By the time they reach 50 they've been a caregiver all their life. They've either had their children or they're parents, and now all of a sudden there's nobody. The husband's left with a bimbo or he's died. Which was my case. And now what do you do with your life? Nobody wants to hire you at that age, even though I think that's the biggest fallacy 111 around. There are lots of people between the age of 50 and 65 that have got more to give than what they realize. There's a lot of hidden talent out there (Wanda's interview, March 15,2000). CoPoMo Subgroup Alexa Alexa was a 28 year old single woman on disability benefits. She lived with her parents and brother; her father was disabled and her mother worked for minimum wage. She spoke of how difficult it was for her parents to accept help. My parents don't have the money, we've had to move around a lot and rent and maybe rely on people more than my parents would like to. Or go to agencies, they find that really hard. Even though there's resources to use... they're too embarrassed to go to them (Alexa's interview, March 20, 2000). At various times she also spoke of taking on her parents' worries. She worried because her parents relied on Alexa's contributions to the rent (through her disabihty I benefits). Eventually, she hoped to live on her own and be more independent. Even though I don't have kids, I have to take on my parents' worries, because my dad's on disability, my mom makes minimum wage babysitting, and they have a lot of money problems, they owe lots to income tax, they owe lots of bills, and I take on their stuff, plus my own. And, I want to get living out on my own, but I need them because I just don't have the rent right now to live on my own. I take on theirs [stress], plus I have mine, and then I want to get better, improve my future too. So I'm also worried about them (Alexa, R T M April 26, 2000). Alexa had debihtating arthritis — she had difficulty walking and used a cane. Her arthritis was both hereditary and a consequence of a manual job. Part of it's [arthritis] hereditary and a past job I did I had to be on my feet all the time and I guess I didn't know how much it was going to impact my health 'cause it sort of happened soon after (Alexa's interview, March 20, 2000). She spoke often of wanting to find work "my main goal is to eventually get work." She was ambivalent about the "benefits" of being on disabihty, though she did not like being labeled "disabled" and was concerned that it may prevent her from being hired. 112 I've been out trying to look for work for the last few years, you're labeled under disabled because of all the things, like you can.. .anytime you go through them you have to.. .like it gives you benefits being labeled that but I don't like to label myself like I feel for the most part that my arthritis, it limits me. But it doesn't limit me enough that I feel that I need a label that they give me (Alexa's interview, March 20, 2000). Alexa had not worked for four years and expressed concern about being unemployed for a long period of time. At one meeting in May 2002, she became emotional when speaking about her desire for work but being unable to find a job. Although Alexa could not participate in many W O A W activities because of her arthritis, she attended most W O A W meetings. Often she quiedy took notes and spoke only when addressed. She suggested that her involvement in W O A W provided important skills that she hoped were transferable to a work environment. Katharine Katharine was a 50 year old single woman with two grown children. She grew up in Montreal but had hved most of her adult life in Calgary and Vancouver. As a single parent she had struggled to provide for her children, and now that they were independent (and Hving in Calgary and Montreal), she felt that she had the time and space to focus on herself. I've been single now since '78, 22 years. I've been divorced and never remarried, so I raised my kids myself so I was single, but I was not single. When you have two kids everything goes to them. AU the food in the fridge goes to them; all the money goes to them. So when that's finally over you can breathe. It's a big stage in life. And some women find, well, I'm nobody anymore. I did not have that problem. Because now I take my hfe back (Katharine's interview, March 14, 2000). Katharine had worked as a secretary, a realtor, and had owned her own pottery business. "I've been a realtor; I've had my pottery business. And when I was a realtor I did a lot of business I am resourceful and artistic but it's sometimes hard to have the guts to do it" (Katharine's interview, March 14, 2000). Katharine had also worked as a translator "I've translated legal and very deep engineering and instruction, computer, oil, heavy industrial and everything" (Katharine's interview, June 9, 2000) and as a researcher for a pharmaceutical 113 company: I've worked in a hospital for a year and a half. I worked in pharmacology at Hoffman la R o c h — And I've done a lot of research because I'm able to research and that's why Hoffman la Roch hired me. And I'm bilingual. And I love research. I did a lot of research on specific medications (Katharine's interview, June 9, 2000). After over 25 years of work, seven years ago Katharine lost her management job as a consequence of downsizing. At that time she was 43 years old. She then spent two years looking for work. I used to fax a hundred resumes a week, and get nothing back I've looked for work in anything, everything that's office work. From management to just straight typing. Not even secretary or clerk or whatever. And nobody wanted to hire me. I was too old for them, older than the managers and bosses, but I was only 43 (Katharine's interview, June 9, 2000). After not finding work, Katharine felt that she had been rejected because she was an older woman. You're looked down, there's no room for failure on this planet. When I was a business woman I was in real estate, I was making $90 000 a year. I had guys who wanted to go out and things, and I didn't want any of them, didn't have time for boyfriends. But then you're a magnet; people come to you. Why? 'Cause you're successful. But once you start going down, they disappear in this world people are too judgmental. They typecast people right away, like a fat woman, a slim woman, but men don't have those judgments. It's always women. And women are the first victims of bank system or credit card system or anything that is money. When I went through my divorce with my lawyer, my lawyer agreed that that women are the first victim of everything (Katharine's interview, March 14, 2000). Katharine said that being rejected affected her health, and that after two years of not frnding work, she fell into a deep depression. It does affect your health. Because you're starting to question yourself. It takes quite a while to realize why they truly are rejecting you. You almost have to be a shrink yourself. And that was humihating. That was your life. And then all a sudden, nobody wants you anymore when I was in real estate, I had a lot of guys ask me out. God I've been proposed seven times Some people say excluded, kicked out or whatever— I felt totally rejected by a lot of people. Totally rejected. When I left real estate and I went back to the office work in Calgary, again I was totally rejected from whole bunch of money-orientated people (Katharine's interview, June 9, 2000). It was only two years ago, however, that Katharine was diagnosed with clinical depression. She 114 said "Well a year and half of the depression. And about a year before that I started really going down financially. Before that I was okay. But I couldn't get a job. I couldn't get anything" (Katharine's interview, June 9, 2000). Katharine experienced her depression as fear, confusion, and social isolation. She explained: Fear is the biggest enemy. It's like a person in front of you. It's like an intelligent being in front of you tricking you. It's really it's like it's got it's consciousness. And it's toying with you. It's very fascmating. If you start having fears... agoraphobia... isolation, and confusion (Katharine's interview, June 9, 2000). It seemed that Katharine's unsuccessful attempts at fmding work had a significant effect on her. At a Research Team meeting she suggested that "nobody wanted" older, single, an